Albert Yang, Mei-Lien Pan, Henry Horng-Shing Lu, Chung-Yueh Lien, Da-Wei Wang, Chih-Hsiung Chen, Der-Cherng Tarng, Dau-Ming Niu, Shih-Hwa Chiou, Chun-Ying Wu, Ying-Chou Sun, Shih-Ann Chen, Shuu-Jiun Wang, Wayne Huey-Herng Sheu, Chi-Hung Lin
Unlabelled: The integration of medical open databases with artificial intelligence (AI) technologies marks a transformative era in biomedical research and health care innovation. Over the past 25 years, initiatives like PhysioNet have revolutionized data access, fostering unprecedented levels of collaboration and accelerating medical discoveries. This rise of medical open databases presents challenges, particularly in harmonizing research enablement with patient confidentiality. In response, privacy laws such as the Health Insurance Portability and Accountability Act have been established, and privacy-enhancing technologies have been adopted to maintain this delicate balance. Privacy-enhancing technologies, including differential privacy, secure multiparty computation, and notably, federated learning (FL), have become instrumental in safeguarding personal health information. FL, in particular, represents a significant advancement by enabling the development and training of AI models on decentralized data. In Taiwan, significant strides have been made in aligning with these global data-sharing and privacy standards. We have actively promoted the sharing of medical data through the development of dynamic consent systems. These systems enable individuals to control and adjust their data-sharing preferences, ensuring transparency and continuity of consent in the ever-evolving landscape of digital health. Despite the challenges associated with privacy protections, the benefits, including improved diagnostics and treatment, are substantial. The availability of open databases has notably accelerated AI research, leading to significant advancements in medical diagnostics and treatments. As the landscape of health care research continues to evolve with open science and FL, the role of medical open databases remains crucial in shaping the future of medicine, promising enhanced patient outcomes and fostering a global research community committed to ethical integrity and privacy.
未标记：医学开放数据库与人工智能（AI）技术的融合标志着生物医学研究和卫生保健创新进入了一个变革时代。在过去的25年里，像PhysioNet这样的计划彻底改变了数据访问，促进了前所未有的合作水平，并加速了医学发现。医疗开放数据库的兴起带来了挑战，特别是在协调研究支持与患者保密方面。作为回应，《健康保险流通与责任法案》（Health Insurance Portability and Accountability Act）等隐私法已经确立，隐私增强技术也已被采用，以维持这种微妙的平衡。隐私增强技术，包括差分隐私、安全多方计算，特别是联邦学习（FL），已成为保护个人健康信息的重要工具。特别是FL，通过在分散的数据上开发和训练人工智能模型，代表了一个重大的进步。台湾在与这些全球数据共享和隐私标准保持一致方面取得了重大进展。我们通过开发动态同意系统积极促进医疗数据的共享。这些系统使个人能够控制和调整他们的数据共享偏好，确保在不断变化的数字卫生环境中，同意的透明度和连续性。尽管与隐私保护相关的挑战，包括改进诊断和治疗在内的好处是巨大的。开放数据库的可用性显著加速了人工智能研究，导致医疗诊断和治疗方面的重大进步。随着开放科学和FL在医疗保健研究领域的不断发展，医疗开放数据库在塑造医学的未来、提高患者治疗效果和培养一个致力于道德诚信和隐私的全球研究社区方面的作用仍然至关重要。
{"title":"Assessing the Evolution and Influence of Medical Open Databases on Biomedical Research and Health Care Innovation: A 25-Year Perspective With a Focus on Privacy and Privacy-Enhancing Technologies.","authors":"Albert Yang, Mei-Lien Pan, Henry Horng-Shing Lu, Chung-Yueh Lien, Da-Wei Wang, Chih-Hsiung Chen, Der-Cherng Tarng, Dau-Ming Niu, Shih-Hwa Chiou, Chun-Ying Wu, Ying-Chou Sun, Shih-Ann Chen, Shuu-Jiun Wang, Wayne Huey-Herng Sheu, Chi-Hung Lin","doi":"10.2196/58954","DOIUrl":"10.2196/58954","url":null,"abstract":"<p><strong>Unlabelled: </strong>The integration of medical open databases with artificial intelligence (AI) technologies marks a transformative era in biomedical research and health care innovation. Over the past 25 years, initiatives like PhysioNet have revolutionized data access, fostering unprecedented levels of collaboration and accelerating medical discoveries. This rise of medical open databases presents challenges, particularly in harmonizing research enablement with patient confidentiality. In response, privacy laws such as the Health Insurance Portability and Accountability Act have been established, and privacy-enhancing technologies have been adopted to maintain this delicate balance. Privacy-enhancing technologies, including differential privacy, secure multiparty computation, and notably, federated learning (FL), have become instrumental in safeguarding personal health information. FL, in particular, represents a significant advancement by enabling the development and training of AI models on decentralized data. In Taiwan, significant strides have been made in aligning with these global data-sharing and privacy standards. We have actively promoted the sharing of medical data through the development of dynamic consent systems. These systems enable individuals to control and adjust their data-sharing preferences, ensuring transparency and continuity of consent in the ever-evolving landscape of digital health. Despite the challenges associated with privacy protections, the benefits, including improved diagnostics and treatment, are substantial. The availability of open databases has notably accelerated AI research, leading to significant advancements in medical diagnostics and treatments. As the landscape of health care research continues to evolve with open science and FL, the role of medical open databases remains crucial in shaping the future of medicine, promising enhanced patient outcomes and fostering a global research community committed to ethical integrity and privacy.</p>","PeriodicalId":16337,"journal":{"name":"Journal of Medical Internet Research","volume":"28 ","pages":"e58954"},"PeriodicalIF":6.0,"publicationDate":"2026-01-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12858049/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146093321","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Yingni Liang, Anran Dai, Meiyan Luo, Zhuolian Zheng, Jiayu Shen, Yinhua Su, Zhongyu Li
<p><strong>Background: </strong>Gestational diabetes mellitus (GDM) is a common complication during pregnancy, with its incidence increasing year by year. It poses numerous adverse health effects on both mothers and newborns. Accurate prediction of GDM can significantly improve patient prognosis. In recent years, artificial intelligence (AI) algorithms have been increasingly used in the construction of GDM prediction models. However, there is still no consensus on the most effective algorithm or model.</p><p><strong>Objective: </strong>This study aimed to evaluate and compare the performance of existing GDM prediction models constructed using AI algorithms and propose strategies for enhancing model generalizability and predictive accuracy, thereby providing evidence-based insights for the development of more accurate and effective GDM prediction models.</p><p><strong>Methods: </strong>A comprehensive search was conducted across PubMed, Web of Science, Cochrane Library, EMBASE, Scopus, and OVID, covering publications from the inception of databases to June 1, 2025, to include studies that developed or validated GDM prediction models based on AI algorithms. Study selection, data extraction, and risk of bias assessment using the Prediction Model Risk of Bias Assessment Tool were performed independently by 2 reviewers. A bivariate mixed-effects model was used to summarize sensitivity and specificity and to generate a summary receiver operating characteristic (SROC) curve, calculating area under the curve (AUC). The Hartung-Knapp-Sidik-Jonkman method was further used to adjust for the pooled sensitivity and specificity. Between-study standard deviation (τ) and variance (τ²) were extracted from the bivariate model to quantify absolute heterogeneity. The Deek test was used to evaluate small-study effects among included studies. Additionally, subgroup analysis and meta-regression were conducted to compare the performance differences among algorithms and to explore sources of heterogeneity.</p><p><strong>Results: </strong>Fourteen studies reported on the predictive value for AI algorithms for GDM. After adjustment with the Hartung-Knapp-Sidik-Jonkman method, the pooled sensitivity and specificity were 0.78 (95% CI 0.69-0.86; τ=0.15, τ2=0.02; PI 0.47-1.09) and 0.85 (95% CI 0.78-0.92; τ=0.11, τ2=0.01; PI 0.59-1.11), respectively. The SROC curve showed that the AUC for predicting GDM using AI algorithms was 0.94 (95% CI 0.92-0.96), indicating a strong predictive capability. Deek test (P=.03) and the funnel plot both showed clear asymmetry, suggesting the presence of small-study effects. Subgroup analysis showed that the random forest algorithm exhibited the highest sensitivity (0.83, 95% CI 0.74-0.93), while the extreme gradient boosting algorithm exhibited the highest specificity (0.82, 95% CI 0.77-0.87). Meta-regression further revealed an evaluation in predictive accuracy in prospective study designs (regression coefficient=2.289, P=.001).</p><p><strong>C
背景：妊娠期糖尿病（GDM）是妊娠期常见的并发症，其发病率呈逐年上升趋势。它对母亲和新生儿造成许多不利的健康影响。准确预测GDM可显著改善患者预后。近年来，人工智能（AI）算法越来越多地应用于GDM预测模型的构建。然而，对于最有效的算法或模型仍然没有达成共识。目的：本研究旨在评估和比较现有基于AI算法构建的GDM预测模型的性能，并提出提高模型通用性和预测精度的策略，从而为开发更准确、更有效的GDM预测模型提供循证见解。方法：对PubMed、Web of Science、Cochrane Library、EMBASE、Scopus和OVID进行综合检索，涵盖从数据库建立到2025年6月1日的出版物，包括基于AI算法开发或验证GDM预测模型的研究。研究选择、数据提取和使用预测模型偏倚风险评估工具进行偏倚风险评估由2名审稿人独立完成。采用双变量混合效应模型对敏感性和特异性进行汇总，并生成综合受试者工作特征（SROC）曲线，计算曲线下面积（AUC）。进一步采用Hartung-Knapp-Sidik-Jonkman方法调整综合敏感性和特异性。从双变量模型中提取研究间标准差（τ）和方差（τ²）来量化绝对异质性。Deek检验用于评价纳入研究中的小研究效应。此外，还进行了亚组分析和元回归，以比较不同算法的性能差异，并探索异质性的来源。结果：有14项研究报道了AI算法对GDM的预测价值。经hartung - knap - sidik - jonkman方法校正后，合并敏感性和特异性分别为0.78 （95% CI 0.69-0.86; τ=0.15, τ2=0.02; PI 0.47-1.09）和0.85 （95% CI 0.78-0.92; τ=0.11, τ2=0.01; PI 0.59-1.11）。SROC曲线显示，人工智能算法预测GDM的AUC为0.94 (95% CI 0.92-0.96)，预测能力较强。Deek检验（P=.03）和漏斗图均显示明显的不对称，提示存在小研究效应。亚组分析显示，随机森林算法灵敏度最高（0.83,95% CI 0.74 ~ 0.93），极端梯度增强算法特异性最高（0.82,95% CI 0.77 ~ 0.87）。meta回归进一步揭示了前瞻性研究设计的预测准确性评价（回归系数=2.289，P=.001）。结论：与以往的叙述性综述不同，本系统综述创新性地提供了用于GDM预测的AI算法的比较和定量综合。这建立了一个以证据为基础的框架来指导模型选择，并确定了一个关键的证据缺口。实际应用的关键含义是在临床采用之前证明了本地验证的必要性。因此，未来的工作应侧重于大规模的前瞻性验证研究，以开发临床适用的工具。
{"title":"Predictive Performance of Artificial Intelligence Algorithms for Gestational Diabetes Mellitus in Pregnant Women: Systematic Review and Meta-Analysis.","authors":"Yingni Liang, Anran Dai, Meiyan Luo, Zhuolian Zheng, Jiayu Shen, Yinhua Su, Zhongyu Li","doi":"10.2196/79729","DOIUrl":"10.2196/79729","url":null,"abstract":"<p><strong>Background: </strong>Gestational diabetes mellitus (GDM) is a common complication during pregnancy, with its incidence increasing year by year. It poses numerous adverse health effects on both mothers and newborns. Accurate prediction of GDM can significantly improve patient prognosis. In recent years, artificial intelligence (AI) algorithms have been increasingly used in the construction of GDM prediction models. However, there is still no consensus on the most effective algorithm or model.</p><p><strong>Objective: </strong>This study aimed to evaluate and compare the performance of existing GDM prediction models constructed using AI algorithms and propose strategies for enhancing model generalizability and predictive accuracy, thereby providing evidence-based insights for the development of more accurate and effective GDM prediction models.</p><p><strong>Methods: </strong>A comprehensive search was conducted across PubMed, Web of Science, Cochrane Library, EMBASE, Scopus, and OVID, covering publications from the inception of databases to June 1, 2025, to include studies that developed or validated GDM prediction models based on AI algorithms. Study selection, data extraction, and risk of bias assessment using the Prediction Model Risk of Bias Assessment Tool were performed independently by 2 reviewers. A bivariate mixed-effects model was used to summarize sensitivity and specificity and to generate a summary receiver operating characteristic (SROC) curve, calculating area under the curve (AUC). The Hartung-Knapp-Sidik-Jonkman method was further used to adjust for the pooled sensitivity and specificity. Between-study standard deviation (τ) and variance (τ²) were extracted from the bivariate model to quantify absolute heterogeneity. The Deek test was used to evaluate small-study effects among included studies. Additionally, subgroup analysis and meta-regression were conducted to compare the performance differences among algorithms and to explore sources of heterogeneity.</p><p><strong>Results: </strong>Fourteen studies reported on the predictive value for AI algorithms for GDM. After adjustment with the Hartung-Knapp-Sidik-Jonkman method, the pooled sensitivity and specificity were 0.78 (95% CI 0.69-0.86; τ=0.15, τ2=0.02; PI 0.47-1.09) and 0.85 (95% CI 0.78-0.92; τ=0.11, τ2=0.01; PI 0.59-1.11), respectively. The SROC curve showed that the AUC for predicting GDM using AI algorithms was 0.94 (95% CI 0.92-0.96), indicating a strong predictive capability. Deek test (P=.03) and the funnel plot both showed clear asymmetry, suggesting the presence of small-study effects. Subgroup analysis showed that the random forest algorithm exhibited the highest sensitivity (0.83, 95% CI 0.74-0.93), while the extreme gradient boosting algorithm exhibited the highest specificity (0.82, 95% CI 0.77-0.87). Meta-regression further revealed an evaluation in predictive accuracy in prospective study designs (regression coefficient=2.289, P=.001).</p><p><strong>C","PeriodicalId":16337,"journal":{"name":"Journal of Medical Internet Research","volume":"28 ","pages":"e79729"},"PeriodicalIF":6.0,"publicationDate":"2026-01-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12858046/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146093355","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Lucas Bickmann, Lucas Plagwitz, Antonius Büscher, Lars Eckardt, Julian Varghese
Background: Electrocardiogram (ECG) data constitutes one of the most widely available biosignal data in clinical and research settings, providing critical insights into cardiovascular diseases as well as broader health conditions. Advancements in deep learning demonstrate high performance in diverse ECG classification tasks, ranging from arrhythmia detection to risk prediction for various diseases. However, the widespread adoption of deep learning for ECG analysis faces significant barriers, including the heterogeneity of file formats, restricted access to pretrained model weights, and complex technical workflows for out-of-domain users.
Objective: This study aims to address major bottlenecks in ECG-based deep learning by introducing ExChanGeAI, an open-source, web-based platform designed to offer an integrated, user-friendly platform for ECG data analysis. Our objective is to streamline the entire workflow-from initial data ingestion (regardless of device or format) and intuitive visualization to privacy-preserving model training and task-specific fine-tuning-making advanced ECG deep learning accessible for both clinical researchers and practitioners without machine learning (ML) expertise.
Methods: ExChanGeAI incorporates robust preprocessing modules for various ECG file types, a set of interactive visualization tools for exploratory data analysis, and multiple state-of-the-art deep learning architectures for ECGs. Users can choose to train models from scratch or fine-tune pretrained models using their own datasets, while all computations are performed locally to ensure data privacy. The platform is adaptable for deployment on personal computers as well as scalable to high-performance computing infrastructures. We demonstrate the platform's performance on several clinically relevant classification tasks across 3 external and heterogeneous validation datasets, including a newly curated test set from routine care, evaluating both model generalizability and resource efficiency.
Results: Our experiments show that de novo training with user-provided, task-specific data can outperform a leading foundation model, while requiring substantially fewer parameters and computational resources. The platform enables users to empirically determine the most suitable model for their specific tasks, based on systematic validations, while lowering technical barriers for out-of-domain experts and promoting open research.
Conclusions: ExChanGeAI provides a comprehensive, privacy-aware platform that democratizes access to ECG analysis and model training. By simplifying complex workflows, ExChanGeAI empowers out-of-domain researchers to use state-of-the-art ML on diverse datasets, democratizing the access to ML in the field of ECG data. The platform is available as open-source code under the Massachusetts Institute of Technology (MIT) license.
{"title":"End-to-End Platform for Electrocardiogram Analysis and Model Fine-Tuning: Development and Validation Study.","authors":"Lucas Bickmann, Lucas Plagwitz, Antonius Büscher, Lars Eckardt, Julian Varghese","doi":"10.2196/81116","DOIUrl":"10.2196/81116","url":null,"abstract":"<p><strong>Background: </strong>Electrocardiogram (ECG) data constitutes one of the most widely available biosignal data in clinical and research settings, providing critical insights into cardiovascular diseases as well as broader health conditions. Advancements in deep learning demonstrate high performance in diverse ECG classification tasks, ranging from arrhythmia detection to risk prediction for various diseases. However, the widespread adoption of deep learning for ECG analysis faces significant barriers, including the heterogeneity of file formats, restricted access to pretrained model weights, and complex technical workflows for out-of-domain users.</p><p><strong>Objective: </strong>This study aims to address major bottlenecks in ECG-based deep learning by introducing ExChanGeAI, an open-source, web-based platform designed to offer an integrated, user-friendly platform for ECG data analysis. Our objective is to streamline the entire workflow-from initial data ingestion (regardless of device or format) and intuitive visualization to privacy-preserving model training and task-specific fine-tuning-making advanced ECG deep learning accessible for both clinical researchers and practitioners without machine learning (ML) expertise.</p><p><strong>Methods: </strong>ExChanGeAI incorporates robust preprocessing modules for various ECG file types, a set of interactive visualization tools for exploratory data analysis, and multiple state-of-the-art deep learning architectures for ECGs. Users can choose to train models from scratch or fine-tune pretrained models using their own datasets, while all computations are performed locally to ensure data privacy. The platform is adaptable for deployment on personal computers as well as scalable to high-performance computing infrastructures. We demonstrate the platform's performance on several clinically relevant classification tasks across 3 external and heterogeneous validation datasets, including a newly curated test set from routine care, evaluating both model generalizability and resource efficiency.</p><p><strong>Results: </strong>Our experiments show that de novo training with user-provided, task-specific data can outperform a leading foundation model, while requiring substantially fewer parameters and computational resources. The platform enables users to empirically determine the most suitable model for their specific tasks, based on systematic validations, while lowering technical barriers for out-of-domain experts and promoting open research.</p><p><strong>Conclusions: </strong>ExChanGeAI provides a comprehensive, privacy-aware platform that democratizes access to ECG analysis and model training. By simplifying complex workflows, ExChanGeAI empowers out-of-domain researchers to use state-of-the-art ML on diverse datasets, democratizing the access to ML in the field of ECG data. The platform is available as open-source code under the Massachusetts Institute of Technology (MIT) license.</p>","PeriodicalId":16337,"journal":{"name":"Journal of Medical Internet Research","volume":"28 ","pages":"e81116"},"PeriodicalIF":6.0,"publicationDate":"2026-01-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12858047/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146093337","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
<p><strong>Background: </strong>Mobile health (mHealth) apps target diverse health behaviors, but engagement may vary by purpose.</p><p><strong>Objective: </strong>This study examined the prevalence, usage patterns, and user characteristics of mHealth apps among Czech adults with internet access, focusing on sociodemographics, digital knowledge and use, and health indicators predicting wellness- and illness-related app use.</p><p><strong>Methods: </strong>Overall, 4775 Czech adults (2365/4775, 49.53% women) aged 18-95 (mean 45.37, SD 16.40) years completed an online survey. Sociodemographic factors included age, gender, education, and income. Digital knowledge and use were measured using the eHealth Literacy Scale and the passive/active use of social networking sites (SNS) for health information. Health indicators covered symptom severity, physical activity, BMI, and eating disorder-related risk propensity (body dissatisfaction, dietary restraint, and weight/shape overvaluation). Participants reported app use for sports, number of steps, nutrition, vitals, sleep, diagnosed conditions, reproductive health, diagnosis assistance, mood and mental well-being, and emergency care guidance. Multivariate hierarchical binary logistic regression analysis identified characteristics of app users. Exploratory structural equation modeling (ESEM) clustered apps into "promoting wellness" and "managing illness" and examined the predictors of frequency of use.</p><p><strong>Results: </strong>Of 4440 respondents, 2172 (48.92%) used mHealth apps. Users were younger (odds ratio [OR] 0.98, 95% CI 0.98-0.99, P<.001), had a monthly income more than 50,000 CZK (1 CZK=US $0.048; vs ≤20,000 CZK: OR 0.54, 95% CI 0.41-0.7, P<.001; 20,001-35,000 CZK: OR 0.78, 95% CI 0.65-0.93, P=.006; 35,001-50,000 CZK: OR 0.83, 95% CI 0.7-0.99, P=.03), were more eHealth literate (OR 1.17, 95% CI 1.06-1.3, P=.003), used SNS passively for health information (OR 1.35, 95% CI 1.21-1.51, P<.001), and had higher eating disorder risk (OR 1.18, 95% CI 1.12-1.25, P<.001) and physical activity (OR 1.18, 95% CI 1.13-1.23, P<.001) than nonusers. Step-counting apps were most common; 65.99% (1430/2167) used them daily or several times a day, followed by apps for sleep (691/2163, 31.95%), vitals (611/2165, 28.22%), and sports (407/2158, 18.86%). ESEM confirmed a 2-factor structure ("promoting wellness" and "managing illness"; χ²<sub>26</sub>=71.9, comparative fit index=0.99, Tucker-Lewis index=0.99, root-mean-square error of approximation=0.03, and standardized root-mean-square residual=0.03). Frequent use of wellness apps was associated with younger age (standardized β=-0.22, P<.001), higher eHealth literacy (standardized β=0.10, P<.001), and physical activity (standardized β=0.15, P<.001). Illness-management app use was associated with active use of SNS for health information (standardized β=0.62, P<.001) and eating disorder risk (standardized β=0.11, P<.001). Digital knowledge, digital use, and health in
背景：移动健康（mHealth）应用程序针对不同的健康行为，但参与可能因目的而异。目的：本研究调查了捷克成年人互联网接入中移动健康应用的流行程度、使用模式和用户特征，重点关注社会人口统计学、数字知识和使用，以及预测健康和疾病相关应用使用的健康指标。方法：共有4775名年龄在18-95岁（平均45.37岁，标准差16.40岁）的捷克成年人（2365/4775人，女性49.53%）完成了在线调查。社会人口因素包括年龄、性别、教育程度和收入。使用电子健康素养量表和被动/主动使用社交网站（SNS）获取健康信息来衡量数字知识和使用情况。健康指标包括症状严重程度、身体活动、身体质量指数和饮食失调相关的风险倾向（身体不满意、饮食限制和体重/体型高估）。参与者报告了应用程序在运动、步数、营养、生命体征、睡眠、诊断状况、生殖健康、诊断协助、情绪和心理健康以及紧急护理指导方面的使用情况。多元层次二元逻辑回归分析确定了应用程序用户的特征。探索性结构方程模型（ESEM）将应用程序分为“促进健康”和“管理疾病”，并检查了使用频率的预测因子。结果：在4440名受访者中，2172名（48.92%）使用移动健康应用程序。使用者较年轻（比值比[OR] 0.98, 95% CI 0.98-0.99, P26=71.9，比较拟合指数=0.99，Tucker-Lewis指数=0.99，近似均方根误差=0.03，标准化均方根残差=0.03）。频繁使用健康应用程序与年龄更小相关（标准化β=-0.22）。结论：移动健康应用程序的参与反映了更广泛的社会、数字和心理不平等，而不仅仅是个人偏好。鼓励数字包容并解决与身体形象和饮食相关的使用问题，可能有助于确保移动医疗技术不会加剧年龄和用户群体之间现有的健康不平等。
{"title":"Patterns and Characteristics of Mobile App Use to Promote Wellness and Manage Illness: Cross-Sectional Study.","authors":"Hayriye Gulec, David Smahel, Yi Huang","doi":"10.2196/71363","DOIUrl":"https://doi.org/10.2196/71363","url":null,"abstract":"<p><strong>Background: </strong>Mobile health (mHealth) apps target diverse health behaviors, but engagement may vary by purpose.</p><p><strong>Objective: </strong>This study examined the prevalence, usage patterns, and user characteristics of mHealth apps among Czech adults with internet access, focusing on sociodemographics, digital knowledge and use, and health indicators predicting wellness- and illness-related app use.</p><p><strong>Methods: </strong>Overall, 4775 Czech adults (2365/4775, 49.53% women) aged 18-95 (mean 45.37, SD 16.40) years completed an online survey. Sociodemographic factors included age, gender, education, and income. Digital knowledge and use were measured using the eHealth Literacy Scale and the passive/active use of social networking sites (SNS) for health information. Health indicators covered symptom severity, physical activity, BMI, and eating disorder-related risk propensity (body dissatisfaction, dietary restraint, and weight/shape overvaluation). Participants reported app use for sports, number of steps, nutrition, vitals, sleep, diagnosed conditions, reproductive health, diagnosis assistance, mood and mental well-being, and emergency care guidance. Multivariate hierarchical binary logistic regression analysis identified characteristics of app users. Exploratory structural equation modeling (ESEM) clustered apps into \"promoting wellness\" and \"managing illness\" and examined the predictors of frequency of use.</p><p><strong>Results: </strong>Of 4440 respondents, 2172 (48.92%) used mHealth apps. Users were younger (odds ratio [OR] 0.98, 95% CI 0.98-0.99, P<.001), had a monthly income more than 50,000 CZK (1 CZK=US $0.048; vs ≤20,000 CZK: OR 0.54, 95% CI 0.41-0.7, P<.001; 20,001-35,000 CZK: OR 0.78, 95% CI 0.65-0.93, P=.006; 35,001-50,000 CZK: OR 0.83, 95% CI 0.7-0.99, P=.03), were more eHealth literate (OR 1.17, 95% CI 1.06-1.3, P=.003), used SNS passively for health information (OR 1.35, 95% CI 1.21-1.51, P<.001), and had higher eating disorder risk (OR 1.18, 95% CI 1.12-1.25, P<.001) and physical activity (OR 1.18, 95% CI 1.13-1.23, P<.001) than nonusers. Step-counting apps were most common; 65.99% (1430/2167) used them daily or several times a day, followed by apps for sleep (691/2163, 31.95%), vitals (611/2165, 28.22%), and sports (407/2158, 18.86%). ESEM confirmed a 2-factor structure (\"promoting wellness\" and \"managing illness\"; χ²<sub>26</sub>=71.9, comparative fit index=0.99, Tucker-Lewis index=0.99, root-mean-square error of approximation=0.03, and standardized root-mean-square residual=0.03). Frequent use of wellness apps was associated with younger age (standardized β=-0.22, P<.001), higher eHealth literacy (standardized β=0.10, P<.001), and physical activity (standardized β=0.15, P<.001). Illness-management app use was associated with active use of SNS for health information (standardized β=0.62, P<.001) and eating disorder risk (standardized β=0.11, P<.001). Digital knowledge, digital use, and health in","PeriodicalId":16337,"journal":{"name":"Journal of Medical Internet Research","volume":"28 ","pages":"e71363"},"PeriodicalIF":6.0,"publicationDate":"2026-01-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146093388","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}
Giuliano Duarte-Anselmi, Susana Sanduvete-Chaves, Salvador Chacón-Moscoso, Daniel López-Arenas
<p><strong>Background: </strong>Unsafe sexual practices remain a major contributor to global morbidity, premature mortality, and health care burden. More than 1 million people acquire a sexually transmitted infection (STI) daily, including HIV. Although biomedical innovations such as pre-exposure prophylaxis have expanded prevention options, consistent condom use and regular HIV and STI testing remain essential behavioral strategies. Adherence to these behaviors remains uneven, underscoring the need for complementary digital and behavioral approaches. Digital behavior change interventions (DBCIs), technology-based programs designed to support health-related behavior change, offer scalable and personalized tools for safer-sex promotion. However, evidence regarding their behavioral components and effectiveness remains fragmented across systematic reviews (SRs).</p><p><strong>Objective: </strong>This study aims to synthesize and critically appraise evidence on the effectiveness of DBCIs for preventing STIs and HIV, and to identify which behavior change techniques (BCTs) and theoretical domains framework (TDF) have been used to improve safe-sex behaviors.</p><p><strong>Methods: </strong>A search was conducted in MEDLINE, Cochrane Database of SRs, Epistemonikos, and PsycINFO for all publications up to November 12, 2025, without language or date restrictions. Eligible SRs examined DBCIs targeting STI and HIV prevention or reduction of risky sexual behaviors. Two reviewers (GDA and DLA) independently screened, extracted data, and appraised methodological quality using the AMSTAR-2 tool. The reporting followed the PRIOR (Preferred Reporting Items for Overviews of Reviews) and PRISMA-S (Preferred Reporting Items for SRs and Meta-Analyses Literature Search Extension) recommendations.</p><p><strong>Results: </strong>Overall, 23 SRs, comprising 514 primary studies and 129,481 participants, met the inclusion criteria. Most interventions were SMS-based, mobile app-based, or web-delivered. Digital interventions consistently improved STI and HIV testing uptake and engagement with sexual health services. Evidence for condom use and biological outcomes was mixed. Improvements in cognitive determinants, such as HIV-related knowledge, motivation, and self-efficacy, were frequently reported. Only 4 reviews explicitly applied BCT or TDF taxonomies, identifying goal setting, feedback on behavior, and prompts and cues as commonly used techniques. Research predominantly originated from high-income settings, with limited evidence from low- and middle-income countries and minimal reporting of sex- or gender-disaggregated outcomes.</p><p><strong>Conclusions: </strong>DBCIs show promise for strengthening STI/HIV prevention, particularly by increasing testing behaviors and supporting cognitive determinants of risk reduction. However, sustained condom use and biological outcomes remain inconsistent, and reporting of behavioral mechanisms is limited. This overview is the first
{"title":"Behavioral Determinants and Effectiveness of Digital Behavior Change Interventions for the Prevention of Sexually Transmitted Infections and HIV: Overview of Systematic Reviews.","authors":"Giuliano Duarte-Anselmi, Susana Sanduvete-Chaves, Salvador Chacón-Moscoso, Daniel López-Arenas","doi":"10.2196/74201","DOIUrl":"10.2196/74201","url":null,"abstract":"<p><strong>Background: </strong>Unsafe sexual practices remain a major contributor to global morbidity, premature mortality, and health care burden. More than 1 million people acquire a sexually transmitted infection (STI) daily, including HIV. Although biomedical innovations such as pre-exposure prophylaxis have expanded prevention options, consistent condom use and regular HIV and STI testing remain essential behavioral strategies. Adherence to these behaviors remains uneven, underscoring the need for complementary digital and behavioral approaches. Digital behavior change interventions (DBCIs), technology-based programs designed to support health-related behavior change, offer scalable and personalized tools for safer-sex promotion. However, evidence regarding their behavioral components and effectiveness remains fragmented across systematic reviews (SRs).</p><p><strong>Objective: </strong>This study aims to synthesize and critically appraise evidence on the effectiveness of DBCIs for preventing STIs and HIV, and to identify which behavior change techniques (BCTs) and theoretical domains framework (TDF) have been used to improve safe-sex behaviors.</p><p><strong>Methods: </strong>A search was conducted in MEDLINE, Cochrane Database of SRs, Epistemonikos, and PsycINFO for all publications up to November 12, 2025, without language or date restrictions. Eligible SRs examined DBCIs targeting STI and HIV prevention or reduction of risky sexual behaviors. Two reviewers (GDA and DLA) independently screened, extracted data, and appraised methodological quality using the AMSTAR-2 tool. The reporting followed the PRIOR (Preferred Reporting Items for Overviews of Reviews) and PRISMA-S (Preferred Reporting Items for SRs and Meta-Analyses Literature Search Extension) recommendations.</p><p><strong>Results: </strong>Overall, 23 SRs, comprising 514 primary studies and 129,481 participants, met the inclusion criteria. Most interventions were SMS-based, mobile app-based, or web-delivered. Digital interventions consistently improved STI and HIV testing uptake and engagement with sexual health services. Evidence for condom use and biological outcomes was mixed. Improvements in cognitive determinants, such as HIV-related knowledge, motivation, and self-efficacy, were frequently reported. Only 4 reviews explicitly applied BCT or TDF taxonomies, identifying goal setting, feedback on behavior, and prompts and cues as commonly used techniques. Research predominantly originated from high-income settings, with limited evidence from low- and middle-income countries and minimal reporting of sex- or gender-disaggregated outcomes.</p><p><strong>Conclusions: </strong>DBCIs show promise for strengthening STI/HIV prevention, particularly by increasing testing behaviors and supporting cognitive determinants of risk reduction. However, sustained condom use and biological outcomes remain inconsistent, and reporting of behavioral mechanisms is limited. This overview is the first ","PeriodicalId":16337,"journal":{"name":"Journal of Medical Internet Research","volume":"28 ","pages":"e74201"},"PeriodicalIF":6.0,"publicationDate":"2026-01-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146086007","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}
Background: Metabolic dysfunction-associated steatohepatitis (MASH) cirrhosis is a leading indication for liver transplantation (LT). Patients with MASH cirrhosis are complex and often have extensive comorbidities. The current model for end-stage liver disease (MELD)-based liver allocation system has suboptimal concordance in predicting waitlist mortality for patients with MASH cirrhosis. Furthermore, it does not capture the competing outcomes of death and LT on the liver transplant waitlist.
Objective: A competing risk analysis using deep learning was conducted to forecast waitlist trajectories of patients with MASH cirrhosis using data available at the time of waitlisting.
Methods: A deep learning competing risk model was constructed using data from 17,551 waitlisted patients with MASH cirrhosis in the Scientific Registry of Transplant Recipients (SRTR) based on the DeepHit model framework with five-fold cross-validation. Model performance was evaluated and compared to single-risk Cox proportional hazards and random survival forests (RSF) models in predicting death or transplant using the concordance index and Brier score. Additionally, a novel performance metric, the competing event coherence (CEC) score, was developed to evaluate model performance in the setting of competing risks. Features associated with death and transplant in the DeepHit model were identified using permutation importance. Models were externally validated on data from the University Health Network.
Results: A total of 17,551 patients were included. The mean MELD at listing was 19.4 (SD 8.1). At 120 months of follow-up on the waitlist, 54.6% (9599/17551) of patients underwent LT, 25.6% (4510/17551) of patients died or were removed due to deterioration, and 19.8% (3442/17551) of patients were removed for improvement or were censored. In a competing risk scenario, DeepHit achieved the best CEC scores at 1 (0.813), 3 (0.811), 6 (0.794), and 12 months (0.772) on the waitlist. The cause-specific RSF model had the highest concordance indices for death or transplant at all time points (death: 0.874 at 1 month, 0.840 at 6 months, and 0.814 at 12 months) except for death at 3 months, where DeepHit (0.883) outperformed RSF. RSF also had lower Brier scores overall, except for transplant at 12 months, where DeepHit outperformed RSF (0.206 vs 0.228). These results were similar on external validation. On feature importance assessment, MELD at listing and its components, as well as functional status, age, and blood type, were associated with death and transplant on the waitlist.
Conclusions: A deep learning competing risk analysis can forecast the risks of both death and transplant in patients with MASH on the waitlist, helping to inform clinical decisions by identifying the most impactful covariates for each outcome.
背景：代谢功能障碍相关脂肪性肝炎（MASH）肝硬化是肝移植（LT）的主要指征。MASH肝硬化患者是复杂的，经常有广泛的合并症。目前基于终末期肝病（MELD）的肝脏分配系统模型在预测MASH肝硬化患者等待名单死亡率方面一致性不佳。此外，它没有捕捉到肝移植等待名单上死亡和肝移植的竞争结果。目的：使用深度学习进行竞争风险分析，利用等待名单时可用的数据预测MASH肝硬化患者的等待名单轨迹。方法：基于深度学习竞争风险模型框架构建深度学习竞争风险模型，使用移植接受者科学登记处（SRTR）中17,551例MASH肝硬化等待患者的数据，并进行五重交叉验证。使用一致性指数和Brier评分评估模型的性能，并将其与单风险Cox比例风险和随机生存森林（RSF）模型在预测死亡或移植方面进行比较。此外，开发了一种新的绩效指标，即竞争事件一致性（CEC）评分，用于评估竞争风险设置下的模型绩效。在DeepHit模型中，与死亡和移植相关的特征使用排列重要性来确定。模型通过来自大学健康网络的数据进行外部验证。结果：共纳入17551例患者。上市时的平均MELD为19.4 （SD 8.1）。在等待名单的120个月随访中，54.6%（9599/17551）的患者接受了肝移植，25.6%（4510/17551）的患者死亡或因恶化而被移除，19.8%（3442/17551）的患者因改善而被移除或被删除。在竞争风险情景中，DeepHit的CEC得分最高，分别为1（0.813）、3（0.811）、6（0.794）和12个月（0.772）。病因特异性RSF模型在所有时间点的死亡或移植的一致性指数最高（1个月死亡：0.874,6个月0.840,12个月0.814），但3个月死亡除外，其中DeepHit（0.883）优于RSF。RSF的Brier评分也较低，但12个月移植时，DeepHit优于RSF （0.206 vs 0.228）。这些结果在外部验证上是相似的。在特征重要性评估中，MELD列表及其组成部分、功能状态、年龄和血型与等待名单上的死亡和移植相关。结论：深度学习竞争风险分析可以预测等待名单上的MASH患者的死亡和移植风险，通过确定每个结果最具影响力的协变量，帮助为临床决策提供信息。
{"title":"Forecasting Waitlist Trajectories for Patients With Metabolic Dysfunction-Associated Steatohepatitis Cirrhosis: A Neural Network Competing Risk Analysis.","authors":"Gopika Punchhi, Yingji Sun, Eunice Tan, Naomi Khaing Than Hlaing, Chang Liu, Sumeet Asrani, Sirisha Rambhatla, Mamatha Bhat","doi":"10.2196/68247","DOIUrl":"10.2196/68247","url":null,"abstract":"<p><strong>Background: </strong>Metabolic dysfunction-associated steatohepatitis (MASH) cirrhosis is a leading indication for liver transplantation (LT). Patients with MASH cirrhosis are complex and often have extensive comorbidities. The current model for end-stage liver disease (MELD)-based liver allocation system has suboptimal concordance in predicting waitlist mortality for patients with MASH cirrhosis. Furthermore, it does not capture the competing outcomes of death and LT on the liver transplant waitlist.</p><p><strong>Objective: </strong>A competing risk analysis using deep learning was conducted to forecast waitlist trajectories of patients with MASH cirrhosis using data available at the time of waitlisting.</p><p><strong>Methods: </strong>A deep learning competing risk model was constructed using data from 17,551 waitlisted patients with MASH cirrhosis in the Scientific Registry of Transplant Recipients (SRTR) based on the DeepHit model framework with five-fold cross-validation. Model performance was evaluated and compared to single-risk Cox proportional hazards and random survival forests (RSF) models in predicting death or transplant using the concordance index and Brier score. Additionally, a novel performance metric, the competing event coherence (CEC) score, was developed to evaluate model performance in the setting of competing risks. Features associated with death and transplant in the DeepHit model were identified using permutation importance. Models were externally validated on data from the University Health Network.</p><p><strong>Results: </strong>A total of 17,551 patients were included. The mean MELD at listing was 19.4 (SD 8.1). At 120 months of follow-up on the waitlist, 54.6% (9599/17551) of patients underwent LT, 25.6% (4510/17551) of patients died or were removed due to deterioration, and 19.8% (3442/17551) of patients were removed for improvement or were censored. In a competing risk scenario, DeepHit achieved the best CEC scores at 1 (0.813), 3 (0.811), 6 (0.794), and 12 months (0.772) on the waitlist. The cause-specific RSF model had the highest concordance indices for death or transplant at all time points (death: 0.874 at 1 month, 0.840 at 6 months, and 0.814 at 12 months) except for death at 3 months, where DeepHit (0.883) outperformed RSF. RSF also had lower Brier scores overall, except for transplant at 12 months, where DeepHit outperformed RSF (0.206 vs 0.228). These results were similar on external validation. On feature importance assessment, MELD at listing and its components, as well as functional status, age, and blood type, were associated with death and transplant on the waitlist.</p><p><strong>Conclusions: </strong>A deep learning competing risk analysis can forecast the risks of both death and transplant in patients with MASH on the waitlist, helping to inform clinical decisions by identifying the most impactful covariates for each outcome.</p>","PeriodicalId":16337,"journal":{"name":"Journal of Medical Internet Research","volume":"28 ","pages":"e68247"},"PeriodicalIF":6.0,"publicationDate":"2026-01-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12854276/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146086050","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
<p><strong>Background: </strong>Online health communities have evolved into digital marketplaces where physicians have to compete for patients. Existing research examines physician-patient dynamics through a patient-centric lens, treating physicians as passive recipients of ratings and reviews, while the strategic role of physician self-disclosure remains unexamined. This gap constrains a comprehensive understanding of how physicians can actively shape patient decisions, making the investigation of strategic self-disclosure imperative.</p><p><strong>Objective: </strong>This study aims to investigate the relationship between physician self-disclosure breadth (scope of information) and depth (detailed expertise) and patient decision-making, as well as whether regional digital health care level (DHL) moderates these relationships.</p><p><strong>Methods: </strong>We conducted a cross-sectional analysis of observational data to test these relationships. Data were collected from China's online health care platform Haodf from September to December 2024. Self-disclosure breadth (including clinical performance, academic experience, and social reputation), self-disclosure depth (including expertise coverage, richness, and granularity), and patient decision-making (total visits) were captured through manual content coding and quantitative measurement. We used structured content analysis to extract the disclosure components, informational scope, and descriptive details of each profile. Then, using validated operational formulas, we calculated the composite indices for disclosure breadth and depth based on the coded dimensions. The study generated 1798 final physician samples with complete data across 14 focal variables. The hypotheses were tested using an ordinary least squares regression model, and 4 robustness checks were conducted, including variable substitution and different resampling techniques.</p><p><strong>Results: </strong>In the primary ordinary least squares regression models, self-disclosure breadth was significantly and positively associated with patient visits (β=0.255, 95% CI 0.054-0.456; P=.01), as was self-disclosure depth (β=0.098, 95% CI 0.030-0.167; P=.005). The breadth×DHL interaction was positive and significant (β=0.261, 95% CI 0.061-0.461; P=.01). Similarly, the depth×DHL interaction was positive and significant (β=0.070, 95% CI 0.002-0.138; P=.045). It should be noted that the association for self-disclosure breadth was stronger than that of self-disclosure depth. DHL strengthened the relationship between the disclosure strategies with patient visits. This contextual amplification indicates that DHL serves as a critical boundary condition, determining the degree to which physician self-disclosure strategies translate into patient acquisition outcomes.</p><p><strong>Conclusions: </strong>This study reconceptualizes physicians as strategic agents shaping patient decision-making through purposeful self-disclosure. Different from exist
背景：在线健康社区已经演变成数字市场，医生们必须为病人竞争。现有的研究通过以患者为中心的视角来审视医患动态，将医生视为评级和评论的被动接受者，而医生自我披露的战略作用仍未得到检验。这一差距限制了对医生如何积极塑造患者决策的全面理解，使得战略自我披露的调查势在必行。目的：本研究旨在探讨医师自我披露广度（信息范围）和深度（详细专业知识）与患者决策的关系，以及区域数字医疗水平（DHL）是否调节了这些关系。方法：我们对观察数据进行了横断面分析，以检验这些关系。数据于2024年9月至12月从中国在线医疗平台浩特收集。通过人工内容编码和定量测量获取自我披露广度（包括临床表现、学术经历和社会声誉）、自我披露深度（包括专业知识覆盖、丰富度和粒度）和患者决策（总访问量）。我们使用结构化内容分析来提取每个概要文件的公开组件、信息范围和描述性细节。然后，利用经过验证的运算公式，基于编码维度计算披露广度和披露深度的综合指数。该研究产生了1798个最终医生样本，其中包含14个焦点变量的完整数据。使用普通最小二乘回归模型对假设进行检验，并进行4次稳健性检验，包括变量替换和不同的重采样技术。结果：在初级普通最小二乘回归模型中，自我披露广度与患者就诊呈显著正相关（β=0.255, 95% CI 0.054 ~ 0.456， P= 0.01），自我披露深度与患者就诊呈显著正相关（β=0.098, 95% CI 0.030 ~ 0.167， P= 0.005）。breadth×DHL交互作用为正且显著（β=0.261, 95% CI 0.061 ~ 0.461； P= 0.01）。同样，depth×DHL相互作用为正且显著（β=0.070, 95% CI 0.002-0.138； P= 0.045）。值得注意的是，自我表露广度的相关性强于自我表露深度。DHL加强了信息披露策略与患者访问量之间的关系。这种背景放大表明DHL是一个关键的边界条件，决定了医生自我披露策略转化为患者获得结果的程度。结论：本研究将医生重新定义为通过有目的的自我披露来塑造患者决策的战略代理人。与现有研究将医生视为评级和评论的被动接受者不同，我们的研究表明，医生可以通过自我披露的广度和深度战略性地塑造患者获得。本研究通过证明自我披露是一种可行的患者获取机制，其中DHL是一个关键的边界条件，为数字健康市场带来了新的见解。研究结果具有现实意义：(1)医生可以利用基于证据的信息披露策略；(2)平台应该实施情境适应性特征；(3)政策制定者应该优先考虑数字基础设施投资，以提高医生的竞争力和患者决策质量。
{"title":"The Relationship Between Physician Self-Disclosure and Patient Acquisition in Digital Health Markets: Cross-Sectional Study.","authors":"Quanchen Liu, Pengqing Yin, Jing Fan","doi":"10.2196/84963","DOIUrl":"https://doi.org/10.2196/84963","url":null,"abstract":"<p><strong>Background: </strong>Online health communities have evolved into digital marketplaces where physicians have to compete for patients. Existing research examines physician-patient dynamics through a patient-centric lens, treating physicians as passive recipients of ratings and reviews, while the strategic role of physician self-disclosure remains unexamined. This gap constrains a comprehensive understanding of how physicians can actively shape patient decisions, making the investigation of strategic self-disclosure imperative.</p><p><strong>Objective: </strong>This study aims to investigate the relationship between physician self-disclosure breadth (scope of information) and depth (detailed expertise) and patient decision-making, as well as whether regional digital health care level (DHL) moderates these relationships.</p><p><strong>Methods: </strong>We conducted a cross-sectional analysis of observational data to test these relationships. Data were collected from China's online health care platform Haodf from September to December 2024. Self-disclosure breadth (including clinical performance, academic experience, and social reputation), self-disclosure depth (including expertise coverage, richness, and granularity), and patient decision-making (total visits) were captured through manual content coding and quantitative measurement. We used structured content analysis to extract the disclosure components, informational scope, and descriptive details of each profile. Then, using validated operational formulas, we calculated the composite indices for disclosure breadth and depth based on the coded dimensions. The study generated 1798 final physician samples with complete data across 14 focal variables. The hypotheses were tested using an ordinary least squares regression model, and 4 robustness checks were conducted, including variable substitution and different resampling techniques.</p><p><strong>Results: </strong>In the primary ordinary least squares regression models, self-disclosure breadth was significantly and positively associated with patient visits (β=0.255, 95% CI 0.054-0.456; P=.01), as was self-disclosure depth (β=0.098, 95% CI 0.030-0.167; P=.005). The breadth×DHL interaction was positive and significant (β=0.261, 95% CI 0.061-0.461; P=.01). Similarly, the depth×DHL interaction was positive and significant (β=0.070, 95% CI 0.002-0.138; P=.045). It should be noted that the association for self-disclosure breadth was stronger than that of self-disclosure depth. DHL strengthened the relationship between the disclosure strategies with patient visits. This contextual amplification indicates that DHL serves as a critical boundary condition, determining the degree to which physician self-disclosure strategies translate into patient acquisition outcomes.</p><p><strong>Conclusions: </strong>This study reconceptualizes physicians as strategic agents shaping patient decision-making through purposeful self-disclosure. Different from exist","PeriodicalId":16337,"journal":{"name":"Journal of Medical Internet Research","volume":"28 ","pages":"e84963"},"PeriodicalIF":6.0,"publicationDate":"2026-01-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146086180","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}
Andrea Nederveld, Elise A Robertson, Angela M Lanigan, Elisabeth F Callen, Tarin L Clay, Ben Fehnert, Lambros Chrones, Michael L Martin, Margaret McCue, Christina M Hester, Melissa K Filippi
<p><strong>Background: </strong>Depression is pervasive, and rates are rising in the United States. Most people with depression receive care from primary care clinicians, but gaps in the quality of care exist. Team-based approaches to depression care have been shown to aid in treatment and management; yet, challenges exist in implementation. Digital health apps have been shown to be effective in improving depression symptoms and enhancing patient engagement in some populations. Many, however, do not share data with clinical care teams.</p><p><strong>Objective: </strong>This study aimed to understand the barriers to and facilitators for implementation of a digital health program that supports coordinated use by clinical care teams and patients, via a mobile app and care team-facing web interface, for depression in primary care.</p><p><strong>Methods: </strong>This study was part of a larger intervention study that included 4 primary care practices: 2 intervention and 2 control sites. The intervention sites used a patient-facing mobile app and a care team-facing web interface, and the control sites continued usual care. The study team conducted interviews from May to October 2021. Patient and care team participants were recruited toward the end of their study involvement. Separate semistructured interview guides were developed for patient and care team participants. Interviews were recorded and transcribed. Data were coded using Atlas.ti.9, and data analysis was completed using a grounded theory approach.</p><p><strong>Results: </strong>Interviews with patient (n=8) and care team (n=8) participants revealed 3 main topics for program implementation: app/interface usability, tracking, and program recommendations. For app/interface usability, overall, navigation for both patient and care team participants was simple and straightforward. Although app content was relevant, patient participants desired additional educational resources and information to aid in their depression treatment and management. In terms of tracking, care team participants indicated that data obtained via the interface enabled them to monitor patients in between visits; and in some circumstances, these data facilitated conversations with patients about treatment plans. Tracking medication adherence differed among patient participants due to established routines of taking medications consistently, lack of motivation to track, or lack of interest in tracking. Care team participants reported the ability to respond more quickly to side effects. Patients commented on tracking difficulties: confusing response options, insufficient goal attainment response options, not being able to provide details or write notes, and no ability to change goals. Some patient and care team participants perceived that tracking encouraged communication with one another; others perceived tracking as having no impact on shared decision-making.</p><p><strong>Conclusions: </strong>Results suggest implementation
{"title":"Patient and Care Team Perspectives of Barriers to and Facilitators for the Implementation of a Digital Health Program for Depression in Primary Care: Qualitative Study.","authors":"Andrea Nederveld, Elise A Robertson, Angela M Lanigan, Elisabeth F Callen, Tarin L Clay, Ben Fehnert, Lambros Chrones, Michael L Martin, Margaret McCue, Christina M Hester, Melissa K Filippi","doi":"10.2196/72003","DOIUrl":"10.2196/72003","url":null,"abstract":"<p><strong>Background: </strong>Depression is pervasive, and rates are rising in the United States. Most people with depression receive care from primary care clinicians, but gaps in the quality of care exist. Team-based approaches to depression care have been shown to aid in treatment and management; yet, challenges exist in implementation. Digital health apps have been shown to be effective in improving depression symptoms and enhancing patient engagement in some populations. Many, however, do not share data with clinical care teams.</p><p><strong>Objective: </strong>This study aimed to understand the barriers to and facilitators for implementation of a digital health program that supports coordinated use by clinical care teams and patients, via a mobile app and care team-facing web interface, for depression in primary care.</p><p><strong>Methods: </strong>This study was part of a larger intervention study that included 4 primary care practices: 2 intervention and 2 control sites. The intervention sites used a patient-facing mobile app and a care team-facing web interface, and the control sites continued usual care. The study team conducted interviews from May to October 2021. Patient and care team participants were recruited toward the end of their study involvement. Separate semistructured interview guides were developed for patient and care team participants. Interviews were recorded and transcribed. Data were coded using Atlas.ti.9, and data analysis was completed using a grounded theory approach.</p><p><strong>Results: </strong>Interviews with patient (n=8) and care team (n=8) participants revealed 3 main topics for program implementation: app/interface usability, tracking, and program recommendations. For app/interface usability, overall, navigation for both patient and care team participants was simple and straightforward. Although app content was relevant, patient participants desired additional educational resources and information to aid in their depression treatment and management. In terms of tracking, care team participants indicated that data obtained via the interface enabled them to monitor patients in between visits; and in some circumstances, these data facilitated conversations with patients about treatment plans. Tracking medication adherence differed among patient participants due to established routines of taking medications consistently, lack of motivation to track, or lack of interest in tracking. Care team participants reported the ability to respond more quickly to side effects. Patients commented on tracking difficulties: confusing response options, insufficient goal attainment response options, not being able to provide details or write notes, and no ability to change goals. Some patient and care team participants perceived that tracking encouraged communication with one another; others perceived tracking as having no impact on shared decision-making.</p><p><strong>Conclusions: </strong>Results suggest implementation","PeriodicalId":16337,"journal":{"name":"Journal of Medical Internet Research","volume":"28 ","pages":"e72003"},"PeriodicalIF":6.0,"publicationDate":"2026-01-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12854275/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146086152","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Claire Coumau, Frederic Gaspar, Mehdi Zayene, Elliott Bertrand, Lorenzo Alberio, Christian Lovis, Patrick E Beeler, Fabio Rinaldi, Monika Lutters, Marie-Annick Le Pogam, Chantal Csajka
<p><strong>Background: </strong>Bleeding complications are a major contributor to adverse drug events among older inpatients, particularly in those treated with antithrombotic agents. Timely and accurate detection of bleeding events is essential for improving drug safety surveillance and clinical risk management.</p><p><strong>Objective: </strong>The study aimed to develop and validate automated algorithms for detecting major bleeding (MB) and clinically relevant nonmajor bleeding (CRNMB) events from electronic medical records (EMRs) by combining structured data-based rule models and a natural language processing (NLP) approach, and to evaluate their performance and generalizability against a manually reviewed gold standard and an external dataset.</p><p><strong>Methods: </strong>We conducted a multicenter retrospective study using routinely collected EMR data from 3 Swiss university hospitals. Patients 65 years or older who received at least one antithrombotic agent and were hospitalized between January 2015 and December 2016 were included. To detect MB and CRNMB events, rule-based algorithms were developed using structured data (International Statistical Classification of Diseases, 10th Revision, German Modification [ICD-10-GM] codes, laboratory values, transfusion records, and antihemorrhagic prescriptions), with variables and cutoff values defined according to adapted International Society on Thrombosis and Haemostasis definitions and expert consensus. In parallel, a supervised NLP model was applied to discharge summaries from one hospital. A manual review of 754 EMRs served as the reference standard for internal validation, and the algorithm performance of the structured data algorithms (SDA), NLP, and their combination (SDA+NLP) was evaluated against this manually reviewed gold standard using standard performance metrics. External validation was performed on an independent dataset from the Lausanne University Hospital to assess model robustness and generalizability.</p><p><strong>Results: </strong>Among 36,039 inpatient stays, SDA identified 8.26% (n=2979) as MB and 15.04% (n=5419) as CRNMB cases. ICD-10-GM codes alone detected 28.5% (n=849) of MB and 31.48% (n=1706) of CRNMB cases, while laboratory data contributed most to event detection (n=1994, 66.94% for MB and n=3663, 67.60% for CRNMB). Integrating SDA with NLP improved detection, identifying 12.2% (920/7513) of MB and 27.4% (2062/7513) of CRNMB cases at 1 hospital. The combined model achieved the best performance (sensitivity 0.84, positive predictive value 0.51, F1-score 0.64). External validation on Lausanne University Hospital 2021-2022 data (n=24,054 stays) confirmed the algorithms' reproducibility; the prevalence of MB decreased while CRNMB increased, reflecting evolving clinical practices and antithrombotic use patterns.</p><p><strong>Conclusions: </strong>Our integrated approach, combining SDA with NLP, enhances the detection of hemorrhagic events in older hospitalized patient
{"title":"Detection of Antithrombotic-Related Bleeding in Older Inpatients: Multicenter Retrospective Study Using Structured and Unstructured Electronic Health Record Data.","authors":"Claire Coumau, Frederic Gaspar, Mehdi Zayene, Elliott Bertrand, Lorenzo Alberio, Christian Lovis, Patrick E Beeler, Fabio Rinaldi, Monika Lutters, Marie-Annick Le Pogam, Chantal Csajka","doi":"10.2196/77809","DOIUrl":"10.2196/77809","url":null,"abstract":"<p><strong>Background: </strong>Bleeding complications are a major contributor to adverse drug events among older inpatients, particularly in those treated with antithrombotic agents. Timely and accurate detection of bleeding events is essential for improving drug safety surveillance and clinical risk management.</p><p><strong>Objective: </strong>The study aimed to develop and validate automated algorithms for detecting major bleeding (MB) and clinically relevant nonmajor bleeding (CRNMB) events from electronic medical records (EMRs) by combining structured data-based rule models and a natural language processing (NLP) approach, and to evaluate their performance and generalizability against a manually reviewed gold standard and an external dataset.</p><p><strong>Methods: </strong>We conducted a multicenter retrospective study using routinely collected EMR data from 3 Swiss university hospitals. Patients 65 years or older who received at least one antithrombotic agent and were hospitalized between January 2015 and December 2016 were included. To detect MB and CRNMB events, rule-based algorithms were developed using structured data (International Statistical Classification of Diseases, 10th Revision, German Modification [ICD-10-GM] codes, laboratory values, transfusion records, and antihemorrhagic prescriptions), with variables and cutoff values defined according to adapted International Society on Thrombosis and Haemostasis definitions and expert consensus. In parallel, a supervised NLP model was applied to discharge summaries from one hospital. A manual review of 754 EMRs served as the reference standard for internal validation, and the algorithm performance of the structured data algorithms (SDA), NLP, and their combination (SDA+NLP) was evaluated against this manually reviewed gold standard using standard performance metrics. External validation was performed on an independent dataset from the Lausanne University Hospital to assess model robustness and generalizability.</p><p><strong>Results: </strong>Among 36,039 inpatient stays, SDA identified 8.26% (n=2979) as MB and 15.04% (n=5419) as CRNMB cases. ICD-10-GM codes alone detected 28.5% (n=849) of MB and 31.48% (n=1706) of CRNMB cases, while laboratory data contributed most to event detection (n=1994, 66.94% for MB and n=3663, 67.60% for CRNMB). Integrating SDA with NLP improved detection, identifying 12.2% (920/7513) of MB and 27.4% (2062/7513) of CRNMB cases at 1 hospital. The combined model achieved the best performance (sensitivity 0.84, positive predictive value 0.51, F1-score 0.64). External validation on Lausanne University Hospital 2021-2022 data (n=24,054 stays) confirmed the algorithms' reproducibility; the prevalence of MB decreased while CRNMB increased, reflecting evolving clinical practices and antithrombotic use patterns.</p><p><strong>Conclusions: </strong>Our integrated approach, combining SDA with NLP, enhances the detection of hemorrhagic events in older hospitalized patient","PeriodicalId":16337,"journal":{"name":"Journal of Medical Internet Research","volume":"28 ","pages":"e77809"},"PeriodicalIF":6.0,"publicationDate":"2026-01-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12854658/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146086079","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Background: Patients with hepatocellular carcinoma (HCC) undergoing transarterial chemoembolization (TACE) experience significant psychological distress, impacting outcomes. While mindfulness-based interventions (MBIs) are beneficial, access is limited. Internet-delivered MBIs (iMBIs) offer an accessible alternative; yet, qualitative understanding of patient experiences with tailored iMBIs for this specific population is lacking.
Objective: This study aimed to explore the facilitators and barriers of patients with HCC post TACE and participated in tailored iMBIs.
Methods: From November 2020 to December 2022, 11 patients with HCC post TACE who had taken part in tailored iMBIs were purposively recruited from a tertiary hospital in Jilin Province. Data were collected through semistructured interviews lasting 30-60 minutes. The interviews were analyzed using conventional content analysis.
Results: Five main categories emerged from the analysis: (1) mindfulness mindset, including acceptance, calmness, and mood improvement; (2) improvement of physical discomfort, such as better sleep, pain relief, reduced gastrointestinal symptoms, and increased activity levels; (3) resistance to mindfulness practice, including perceived lack of effectiveness, unsuitable conditions, equipment limitations, and difficulty concentrating; (4) support and encouragement, involving social support, supervision, and professional guidance; and (5) accessibility and convenience characterized by restoration of life balance and user-friendly features of the practice. Each category encompassed several subcategories reflecting the diverse experiences of participants.
Conclusions: While iMBIs were generally perceived as convenient and accessible, challenges such as equipment limitations were noted. Future implementation should focus on enhancing supportive factors to improve adherence, minimizing barriers, and refining the design and delivery of iMBI programs.
Trial registration: Chinese Clinical Trial Registry ChiCTR1900027976; https://www.chictr.org.cn/showproj.html?proj=46657.
{"title":"Tailored Internet-Delivered Mindfulness-Based Interventions for Patients With Hepatocellular Carcinoma After Transarterial Chemoembolization: Qualitative Study.","authors":"Zengxia Liu, Min Li, Yong Jia, Li Chen","doi":"10.2196/78337","DOIUrl":"https://doi.org/10.2196/78337","url":null,"abstract":"<p><strong>Background: </strong>Patients with hepatocellular carcinoma (HCC) undergoing transarterial chemoembolization (TACE) experience significant psychological distress, impacting outcomes. While mindfulness-based interventions (MBIs) are beneficial, access is limited. Internet-delivered MBIs (iMBIs) offer an accessible alternative; yet, qualitative understanding of patient experiences with tailored iMBIs for this specific population is lacking.</p><p><strong>Objective: </strong>This study aimed to explore the facilitators and barriers of patients with HCC post TACE and participated in tailored iMBIs.</p><p><strong>Methods: </strong>From November 2020 to December 2022, 11 patients with HCC post TACE who had taken part in tailored iMBIs were purposively recruited from a tertiary hospital in Jilin Province. Data were collected through semistructured interviews lasting 30-60 minutes. The interviews were analyzed using conventional content analysis.</p><p><strong>Results: </strong>Five main categories emerged from the analysis: (1) mindfulness mindset, including acceptance, calmness, and mood improvement; (2) improvement of physical discomfort, such as better sleep, pain relief, reduced gastrointestinal symptoms, and increased activity levels; (3) resistance to mindfulness practice, including perceived lack of effectiveness, unsuitable conditions, equipment limitations, and difficulty concentrating; (4) support and encouragement, involving social support, supervision, and professional guidance; and (5) accessibility and convenience characterized by restoration of life balance and user-friendly features of the practice. Each category encompassed several subcategories reflecting the diverse experiences of participants.</p><p><strong>Conclusions: </strong>While iMBIs were generally perceived as convenient and accessible, challenges such as equipment limitations were noted. Future implementation should focus on enhancing supportive factors to improve adherence, minimizing barriers, and refining the design and delivery of iMBI programs.</p><p><strong>Trial registration: </strong>Chinese Clinical Trial Registry ChiCTR1900027976; https://www.chictr.org.cn/showproj.html?proj=46657.</p>","PeriodicalId":16337,"journal":{"name":"Journal of Medical Internet Research","volume":"28 ","pages":"e78337"},"PeriodicalIF":6.0,"publicationDate":"2026-01-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146086107","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}
扫码关注我们
求助内容：
应助结果提醒方式：
京公网安备 11010802042870号