DIGITAL HEALTH最新文献

{"title":"Development, simplification, and prospective methodological exploration of a Chinese endoscopy patient experience scale and a decision support system.","authors":"Chengcheng Gao, Xicheng Chen, Yan Wang, Weijia Dou, Qiongjie Shao, Rui Zhang, Ying Liang, Nan Li, Fang Hu, Yang Zhao, Zhenxiong Liu, Lei Shang","doi":"10.1177/20552076261480890","DOIUrl":"10.1177/20552076261480890","url":null,"abstract":"<p><strong>Background and objective: </strong>Chinese-language patient-reported experience measures (PREMs) for gastrointestinal endoscopy remain limited. We developed and psychometrically evaluated a 54-item Endoscopy Patient Experience Scale (EPES) for Chinese patients; because the field survey had a valid-response rate of 99.28% (965/972), simplification was framed as a prospective methodological exploration rather than as a response to demonstrated burden.</p><p><strong>Methods: </strong>Colonoscopy responses (n = 662) were split by stratified random sampling into an exploration set (n = 463) for exploratory factor analysis and a validation set (n = 199) for confirmatory factor analysis. A 303-patient gastroscopy cohort was used only as a cross-procedure applicability test. A multi-objective black-winged kite algorithm selected Simplified EPES (S-EPES) items to reproduce EPES dimension scores under reliability and item-reduction constraints, and an exploratory decision support system (EPE-DSS) was built from S-EPES and clustering outputs.</p><p><strong>Results: </strong>The EPES showed acceptable structural, convergent and discriminant validity and reliability in the colonoscopy validation set, and the retained structure was compatible with gastroscopy after removal of bowel-preparation dimensions. The 24-item S-EPES achieved per-dimension R² > 0.85 against EPES dimension scores; post-hoc classification yielded a weighted-average F1 of 0.917 in colonoscopy data.</p><p><strong>Conclusion: </strong>The main contribution is the Chinese EPES. S-EPES and EPE-DSS are exploratory prototypes demonstrating score-prediction feasibility, not validated clinical instruments; multicenter validation, comparison with conventional item-reduction methods, and measurement-invariance testing are required.</p>","PeriodicalId":51333,"journal":{"name":"DIGITAL HEALTH","volume":"12 ","pages":"20552076261480890"},"PeriodicalIF":3.9,"publicationDate":"2026-08-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13500960/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148814619","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Understanding patient feedback on online doctor review platforms: Divergent topics, sentiments, and service needs by disease risk level.","authors":"Jingwen Ma, Ruojia Wang, Keming Fan, Rui Lian","doi":"10.1177/20552076261481348","DOIUrl":"10.1177/20552076261481348","url":null,"abstract":"<p><strong>Introduction: </strong>There is currently limited understanding of patient satisfaction with online healthcare platforms by disease risk levels. This study aims to collect online doctor reviews from online healthcare platforms in China to investigate patient satisfaction by disease risk levels, identify the factors influencing satisfaction, and understand patients' service needs.</p><p><strong>Methods: </strong>Data were collected from Haodf.com, resulting in a total of 24,742 low-risk disease patient reviews and 9,821 high-risk disease patient reviews. BERTopic modeling analysis, sentiment analysis, Kano-IPA modeling analysis and statistical analysis were conducted to derive insights.</p><p><strong>Results: </strong>This study identified 13 topics among low-risk patients and 10 topics among high-risk patients. For both low-risk (coef=0.26) and high-risk (coef=0.17) patients, clinical expertise was the top positive factor influencing satisfaction. Poor value for money was the top negative influencing factor for low-risk patient satisfaction, whereas inadequate information response was the top negative influencing factor for high-risk patient satisfaction. The dimension-level analysis suggests a potential moderating role of disease risk in the factors associated with patient satisfaction. The results of the Kano-IPA modeling analysis indicate that service needs differ between low-risk and high-risk patients.</p><p><strong>Conclusion: </strong>This study provides a comprehensive understanding of factors influencing patient satisfaction and patients' service needs in online healthcare service by disease risk levels. The findings offer valuable implications for doctors' service provision, online healthcare platform design and healthcare management.</p>","PeriodicalId":51333,"journal":{"name":"DIGITAL HEALTH","volume":"12 ","pages":"20552076261481348"},"PeriodicalIF":3.9,"publicationDate":"2026-08-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13500988/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148814806","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"An investigation into the simulation sickness with head-mounted displays in people with aphasia.","authors":"Muhammed Intiaz, Rajath Shenoy, Apoorva Pauranik, Girish Menon, Shivani Tiwari, Aparna Pai, Gopee Krishnan","doi":"10.1177/20552076261462754","DOIUrl":"10.1177/20552076261462754","url":null,"abstract":"<p><strong>Background: </strong>In recent times, with the proliferation of technological innovations, head-mounted displays are increasingly used in healthcare, including clinical rehabilitation. Until recently, their use in people with aphasia (an inability to use language for communication following brain damage) was limited. Emerging developments in Mixed Reality, combining augmented and virtual realities, have enabled the use of non-occlusive head-mounted displays in aphasia rehabilitation. However, concerns about the possible adverse effects, such as simulation sickness, arise in people with brain damage and aphasia.</p><p><strong>Objective: </strong>To investigate whether a mixed reality head-mounted device (the MiRAR, deployed using the Microsoft HoloLens) induces simulation sickness in general and, specifically, in people with stroke-induced aphasia.</p><p><strong>Methods: </strong>Both the healthy participants and those with aphasia following brain damage wore a non-occlusive head-mounted display. They interacted with the virtual avatars within a fully immersive Mixed Reality environment. Simulation sickness was assessed after exposure using a standardised self-report tool, the Simulator Sickness Questionnaire (SSQ).</p><p><strong>Results: </strong>Both groups showed only minimal symptoms of simulation sickness. Most importantly, the two groups did not differ significantly across the domains of the simulation sickness scale.</p><p><strong>Conclusion: </strong>People with aphasia did not experience any additional adverse effects compared to the healthy participants. The implications of these findings for clinical and rehabilitation purposes are discussed.</p>","PeriodicalId":51333,"journal":{"name":"DIGITAL HEALTH","volume":"12 ","pages":"20552076261462754"},"PeriodicalIF":3.9,"publicationDate":"2026-08-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13500961/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148814584","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A scoping review of AI in the classification of oral cancer and oral potentially malignant disorders from visible-light photography.","authors":"Charles Goodmaker, Rishi Bhandari, Anwar Tappuni, Tuan D Pham","doi":"10.1177/20552076261481593","DOIUrl":"10.1177/20552076261481593","url":null,"abstract":"<p><strong>Objectives: </strong>To map the scope, methodological characteristics, and clinical task design of artificial intelligence (AI) models applied to visible-light photographs of the oral cavity for classification of oral cancer and oral potentially malignant disorders (OPMD).</p><p><strong>Methods: </strong>A scoping review following the Arksey and O'Malley framework and PRISMA-ScR was conducted. PubMed, Embase, Scopus, and Web of Science were searched from January 2015 to October 2025. Studies were eligible if they applied AI to intraoral photography for lesion detection, classification, or risk stratification. Data were extracted on dataset provenance, ground-truth labelling, model architecture, validation strategy, performance metrics, and reporting completeness.</p><p><strong>Results: </strong>127 studies met inclusion and were categorised by their output classes across a mutually exclusive, five-category framework (four ordered tiers plus a residual category) with a benign-OPMD-malignant disease spectrum. Over half addressed a cancer-versus-non-cancer endpoint only (54.3%, n=69), whilst 27.6% (n=35) preserved OPMD as a discrete output category. Public image datasets were used in 44.1% (n=56) of studies, and fully histologically-confirmed ground-truth labels were available in only 7.9% (n=10). External validation was reported in 8.7% (n=11), patient-level data splitting in 10.2% (n=13), and prospective clinical evaluation in 0% (n=0). ResNet-family architectures were most frequently reported (50.4%, n=64, as primary or comparator model); explainable-AI components appeared in 37.0% (n=47) and multimodal inputs in 12.6% (n=16).</p><p><strong>Conclusion: </strong>Classifying oral cancer and OPMD lesions with visible light photographs is an expanding field of research, but one which reveals space for more robust and clinically grounded training datasets as well as exploration of frontier AI methodologies.</p>","PeriodicalId":51333,"journal":{"name":"DIGITAL HEALTH","volume":"12 ","pages":"20552076261481593"},"PeriodicalIF":3.9,"publicationDate":"2026-08-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13500969/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148814579","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Ethical governance of artificial intelligence in digital health for Indigenous populations: A narrative review and conceptual framework.","authors":"Amal Khan, Alyson S N Bear, Mairyn Rackow, Cassandra Opikokew Wajuntah, Kenneth Lai, Veronica McKinney, John Costa, Ivar Mendez","doi":"10.1177/20552076261481276","DOIUrl":"10.1177/20552076261481276","url":null,"abstract":"<p><p>Artificial intelligence (AI) integration into digital health care delivery is rapidly expanding and is increasingly viewed as a promising approach for improving access to care, strengthening diagnostic decision-making, and addressing persistent health inequities experienced by Indigenous populations. Although AI-enabled digital health applications offer potential benefits such as scalability, personalization, and improved clinical decision support, they also raise critical ethical concerns related to equitable implementation, culturally grounded design, representative data usage accountability, privacy protection, and Indigenous data sovereignty. A coherent ethical governance framework is necessary to guide the responsible development, deployment, and evaluation of AI applications implemented in digital health solutions involving Indigenous communities. A narrative review was conducted to critically assess recent literature (2020-2025) addressing ethical considerations in the integration of AI into digital health care delivery for Indigenous populations. Searches were performed in MEDLINE, PubMed, Web of Science, and Scopus using terms related to artificial intelligence, digital health, Indigenous populations, and ethics, with supplementary grey literature searches conducted to identify policy and implementation reports. Studies were selected based on relevance to AI-enabled digital health initiatives implemented in digital health contexts involving Indigenous communities in Canada, the United States, Australia, New Zealand, and the United Kingdom, and were synthesized using thematic narrative analysis. Six themes denoting ethical domains were identified: (1) co-design of AI technologies, (2) culturally grounded implementation, (3) bias and representative data, (4) accountability, transparency, and intelligibility, (5) privacy and Indigenous data sovereignty, and (6) resource accessibility. These domains function as interconnected components of ethical governance, with participatory governance and Indigenous data sovereignty emerging as foundational cross-cutting principles. The proposed framework is not intended as a pan-Indigenous or one-size-fits-all model. Instead, it is conceptualized as a flexible, distinctions-based ethical governance framework that must be adapted to the specific cultural, geographic, and governance contexts of individual Indigenous communities. Based on this analysis, we propose a conceptual ethical governance framework to support policymakers, clinicians, health systems, digital health developers, as well as Indigenous communities in identifying ethically appropriate safeguards across the lifecycle of AI-enabled digital health initiatives planned or implemented in Indigenous communities.</p>","PeriodicalId":51333,"journal":{"name":"DIGITAL HEALTH","volume":"12 ","pages":"20552076261481276"},"PeriodicalIF":3.9,"publicationDate":"2026-08-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13500953/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148814820","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Prior use and comfort using WhatsApp for health communication among people living with HIV in urban Panama: A cross-sectional study.","authors":"Amanda Gabster, Zhuo Meng, Beren Crim Sabuncu, Lisa B Hightow-Weidman, Carlos Chávez Sáenz, Dixennia Bejarano, Casey D Xavier Hall, Gonzalo Cabezas Talavero, Ana Belén Araúz, Venus Tejada, Nandin Solís, Barbara Delgado, Umed Ibragimov, Ibrahim Yigit, Frank Frankie Y Wong, Adam E Tratner, Isaac K Payton, Eduardo De La Vega-Taboada, Eugenia Flores Millender","doi":"10.1177/20552076261479438","DOIUrl":"10.1177/20552076261479438","url":null,"abstract":"<p><strong>Introduction: </strong>Digital tools, including WhatsApp, are widely used across Panama. These tools may serve as mobile health (mHealth) interventions to maintain patient contact, especially among individuals who have difficulty attending appointments. However, prior use and comfort with WhatsApp-based health communication among People Living with HIV (PLHIV) have not been previously examined in Panama.</p><p><strong>Methods: </strong>During August-October 2024, this cross-sectional study recruited participants from two urban antiretroviral (ART) clinics and three community-based organizations in and near Panama City and Colón. A self-administered questionnaire was used to collect information on prior use and comfort with WhatsApp-based communication with providers and for receiving health information. Descriptive statistics and multivariable regression described associations with sociodemographic, HIV care engagement, substance use, and psychosocial factors.</p><p><strong>Results: </strong>Among 277 participants (median age 37 years, interquartile range 28-45 years), 33.6% had communicated with providers and 47.7% received health information via WhatsApp. Over 80% reported comfort using WhatsApp with a provider and for receiving health information. Discrimination was associated with prior provider communication (AOR=2.129, 90%CI:1.12-4.047) and comfort (AOR=2.324, 90%CI:1.202-4.492). Missed appointments were associated with lower odds of provider communication (AOR=0.567, 90%CI:0.311-1.034). Recent drug use was associated with having received health information (AOR=2.16, 90%CI:1.054-4.465); not yet engaging in care was associated with lower odds of prior use (AOR=0.374, 90%CI:0.145-0.963).</p><p><strong>Conclusions: </strong>Prior use of WhatsApp for health communication was modest, but comfort was high. Associations with discrimination, drug use, and retention in care may indicate that WhatsApp could be relevant for individuals experiencing psychosocial and behavioural vulnerabilities.</p>","PeriodicalId":51333,"journal":{"name":"DIGITAL HEALTH","volume":"12 ","pages":"20552076261479438"},"PeriodicalIF":3.9,"publicationDate":"2026-08-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13498786/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148801473","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Media literacies, platform imaginaries and mental health influencers: How young people respond to mental health content on TikTok.","authors":"Alex Christiansen, Ruth Page, Shioma-Lei Craythorne, Michael Larkin, Paul Crawford, Evangelina Asiedu-Addo, Emma Garavini","doi":"10.1177/20552076261479043","DOIUrl":"10.1177/20552076261479043","url":null,"abstract":"<p><strong>Objective: </strong>Social media is increasingly used as a source of mental health information for young people, with influencers playing a key role in public health communication. However, little is known about how young people engage with influencer mental health content and the media literacy skills they use to evaluate it. The objective of this study is to explore and evaluate the self-reported media literacy skills and beliefs that young people use when they encounter mental health content on TikTok.</p><p><strong>Methods: </strong>The study uses qualitative, thematic analysis to provide an empirical account of how young people interpret mental health content from influencers on TikTok. Research was conducted via eighteen group (n = 73 participants) and individual interviews (n = 13) with eighty-six participants in the United Kingdom. The analysis focused on self-reported media literacy skills and strategies for navigating influencer-led mental health content.</p><p><strong>Results: </strong>The study identifies three contextual challenges in young people's responses to mental health content on TikTok: (I) reliability and relatability act as distinct but complementary criteria used to assess influencers and their mental health content; (II) platform beliefs influence willingness to engage with mental health content; and (III) algorithmically driven platform design hinders critical thinking about mental health content.</p><p><strong>Conclusions: </strong>These findings highlight the sophistication of young people's media literacy skills alongside the significant, contextual barriers, which hinder the use of social media as a channel for mental health support from influencers. Our results suggest that public health campaigns may need to take a more platform-sensitive approach to online communication, acknowledging the ways platform-driven beliefs and design shape how and why young people engage with mental health content.</p>","PeriodicalId":51333,"journal":{"name":"DIGITAL HEALTH","volume":"12 ","pages":"20552076261479043"},"PeriodicalIF":3.9,"publicationDate":"2026-08-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13617548/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149076571","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Bounded control in digital health: Privacy discourse and algorithmic governance in Apple Health.","authors":"Aybike Serttaş, Sema Bulat Demir, Gizem Uyanik Uğur, Oğuzhan Altunkurt","doi":"10.1177/20552076261483385","DOIUrl":"10.1177/20552076261483385","url":null,"abstract":"<p><strong>Objective: </strong>This study examines how user control is constructed, communicated, and experienced in digital health tracking, using Apple Health as a case study. Rather than treating control as either fully empowering or simply illusory, the study conceptualizes it as bounded, relational, and discursively produced within platform environments.</p><p><strong>Methods: </strong>The study uses a qualitative, triangulated design combining critical discourse analysis of Apple's official \"Health App and Privacy\" text with semi-structured interviews with Apple Watch and Apple Health users (n = 18). The analysis compares corporate privacy discourse with users' everyday experiences of metrics, goals, notifications, permissions, and health-related feedback.</p><p><strong>Results: </strong>Apple's privacy discourse frames users as active decision-makers who can manage privacy, sharing, and platform participation. At the same time, this discourse situates user choice within predefined platform architectures, permissions, and data-processing conditions. Interview findings show that users often understand the limits of platform options and engage with Apple Health in pragmatic and informed ways. However, their everyday practices are also shaped by goal-setting, notifications, quantified feedback, and affective responses such as motivation, reassurance, pressure, guilt, and dependence.</p><p><strong>Conclusions: </strong>The findings suggest that user control in digital health should be understood as bounded rather than simply false. Apple Health provides functional choices and privacy settings, but these operate within a wider system of platform-defined options, behavioral prompts, and datafication. The study contributes to digital health and health communication scholarship by showing how privacy discourse, interface design, and self-tracking practices jointly produce structured forms of autonomy under algorithmic governance.</p>","PeriodicalId":51333,"journal":{"name":"DIGITAL HEALTH","volume":"12 ","pages":"20552076261483385"},"PeriodicalIF":3.9,"publicationDate":"2026-08-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13498789/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148801348","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"ChatGPT as a source of surgical information: Evaluation of responses to patient questions on hallux rigidus fusion.","authors":"Kamil Balaban, Mehmet Batu Ertan, Mahmut Kalem","doi":"10.1177/20552076261481386","DOIUrl":"10.1177/20552076261481386","url":null,"abstract":"<p><strong>Background: </strong>Patients are increasingly turning to online resources and artificial intelligence (AI)-based tools to obtain information about orthopedic conditions and surgical options. Large language models, such as ChatGPT, are becoming prominent in patient education; however, their reliability and readability remain uncertain. This study evaluated the quality and readability of responses generated by ChatGPT-4o and ChatGPT-5 to frequently asked patient questions regarding hallux rigidus fusion surgery.</p><p><strong>Methods: </strong>Twenty commonly asked patient questions were compiled and presented to ChatGPT-4o and ChatGPT-5. Readability was assessed using the Flesch-Kincaid Grade Level, Gunning Fog, Coleman-Liau, and Simple Measure of Gobbledygook indices. Quality was evaluated with the DISCERN tool, response accuracy scores, and Journal of the American Medical Association (JAMA) criteria. Interrater agreement was measured using the Intraclass Correlation Coefficient (ICC).</p><p><strong>Results: </strong>ChatGPT-4o generated longer responses (802 vs. 242 words; p<0.001) with slightly higher readability grade levels (10.81 vs. 10.37; p=0.031). Accuracy (2.00 vs. 1.85; p=0.323) and DISCERN scores (49.35 vs. 48.93; p=0.747) showed no significant differences. All responses received a JAMA score of 0 due to the absence of citations, authorship, or transparency indicators. Interrater reliability indicated moderate to good agreement (ICC: 0.68-0.80).</p><p><strong>Conclusion: </strong>ChatGPT-4o and ChatGPT-5 provide generally satisfactory yet non-comprehensive, limited-quality information at a level above tenth-grade regarding hallux rigidus fusion surgery. Although linguistically coherent, responses lack evidence-based detail and individualized guidance. These models may supplement, but cannot replace, expert orthopedic counseling. Ensuring physician oversight and integrating validated, updated clinical content remain essential for safe implementation of AI-generated patient information.</p>","PeriodicalId":51333,"journal":{"name":"DIGITAL HEALTH","volume":"12 ","pages":"20552076261481386"},"PeriodicalIF":3.9,"publicationDate":"2026-08-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13498785/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148801418","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"When end users can build: AI-assisted application development and the new accountability gap in digital health.","authors":"Bo Xie","doi":"10.1177/20552076261481598","DOIUrl":"10.1177/20552076261481598","url":null,"abstract":"<p><p>Generative artificial intelligence (AI) is dramatically changing the division of labor in digital health innovation. Until recently, a frontline health professional, educator or researcher who identified a workflow problem typically needed a software developer to translate that problem into a functioning application. Drawing on my recent experience building and deploying two web applications for health professional education, each in less than five hours and without formal computer science or information technology training, this Perspective argues that the early bottleneck in digital health tool development is shifting. AI did not replace domain judgment: the clinical education needs, workflow priorities, and decisions about what counted as useful remained mine. What AI supplied was technical translation, code generation, troubleshooting and step-by-step guidance for using tools such as Visual Studio Code, PowerShell, version control and web deployment platforms. This experience suggests that end users may increasingly be able to move from problem identification to first functioning application without a software developer at the initial build stage. The implications are substantial but not simple. Technical collaboration remains essential, but its center of gravity may move toward cybersecurity, privacy, architecture, institutional integration, maintenance and long-term stewardship. The unresolved issue is not whether end users can now build, but who is responsible for determining whether AI-generated code is safe enough to deploy in institutional environments. Digital health practitioners and researchers therefore need governance models that recognize a new reality: end-user-developed applications are becoming technically feasible, but accountable review and ownership remain indispensable.</p>","PeriodicalId":51333,"journal":{"name":"DIGITAL HEALTH","volume":"12 ","pages":"20552076261481598"},"PeriodicalIF":3.9,"publicationDate":"2026-08-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13498793/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148801087","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}