Background: Prior microbiology results, resistance patterns, and antimicrobial exposure are central to safe and effective antimicrobial prescribing. Digital health fragmentation refers to the dispersal of patient data across multiple electronic systems and the associated challenge of accessing complete information at the point of care. Antimicrobial prescribing for infections represents a critical use case to investigate the impact of digital health fragmentation on patient care. While interoperability has been studied in the context of patient safety, no review has described digital health fragmentation within the United Kingdom and examined its impact on antimicrobial prescribing and antimicrobial stewardship (AMS). Objective: This study aimed to (1) characterize the extent of digital health fragmentation in the United Kingdom, (2) summarize the available evidence on its impact on AMS and prescribing practices in high-income countries, and (3) identify potential solutions. Methods: A rapid review of the peer-reviewed literature was conducted following published guidance for rapid reviews and the PRISMA (Preferred Reporting Items of Systematic Reviews and Meta-Analyses) statement. MEDLINE ALL and PsycInfo were searched on August 19, 2025, using search terms relating to digital health fragmentation or interoperability, patient safety, and antimicrobial use. Searches were limited to English-language publications from 2015 (for characterizing the recent trends or current state of digital health fragmentation in the United Kingdom) or 2010 onward (for AMS-related impacts and solutions). Screening was conducted by 4 researchers following predefined inclusion and exclusion criteria. Extracted data were synthesized narratively through framework analysis. Study quality was appraised using the Mixed Methods Appraisal Tool. Results: Fourteen studies met the inclusion criteria. Ten studies described the extent and nature of digital health fragmentation in the United Kingdom. Digital health fragmentation affects a large number of patients and is linked to clinical care efficiency, quality, and safety risks, including limited access to external clinical records, missing or incomplete information, duplicate investigations, delays in decision‑making, and substantial time spent searching for data. Evidence specific to antimicrobial prescribing was limited (4 studies) but indicated that AMS relies on information spread across multiple systems, with poor interoperability disrupting workflows, hindering communication, and undermining stewardship activities. Only 1 study reported the development of a digital tool designed to address digital health fragmentation and support AMS. Conclusions: Digital health fragmentation negatively affects patient care across the United Kingdom, yet evidence on how it impacts AMS remains scarce. Given the urgency of the global antimicrobial resistance crisis, future research should therefore quantify the scale and impact of digital health fragmentation for AMS to inform investment and innovation in digital infrastructure and clinical-supportive solutions. Trial Registration: PROSPERO CRD420251126067; https://www.crd.york.ac.uk/PROSPERO/view/CRD420251126067

Background: Assessing medication adherence is central to quality care, yet linking electronic health record (EHR) medication orders to outpatient pharmacy dispense data remains technically complex. Objective: This study aimed to present a generalized, reproducible tutorial for linking EHR medication orders to pharmacy dispense data that can be used to assess medication dispense proportions. Methods: We developed and validated a structured query approach to link EHR medication orders to external pharmacy dispense data using patient identifiers, medication-level identifiers, pharmacy identifiers, and temporal constraints. The tutorial emphasizes key design decisions, including handling multiple triggering events, deduplication across vendors, and managing formulation changes. A retrospective cohort of pediatric acute otitis media encounters (January 1, 2021, to January 1, 2024) was used as an illustrative example. Results: Overall, 98.3% (302/307) of pharmacies in the cohort returned at least 1 dispense record during the study period and were therefore classified as reporting pharmacies. Among 3404 orders, 2616 (76.9%) had a recorded dispense. Conclusions: EHR-integrated pharmacy data provide a feasible, timely proxy for assessing medication adherence. This tutorial provides a scalable framework for linking EHR and pharmacy data for medication adherence studies, while highlighting key methodological considerations for SQL coding.

Background: Family caregivers of children with chronic health conditions experience substantial physical and mental health burdens, including burnout, anxiety, depression, fatigue, and sleep disturbances. Despite this need, validated digital mental health tools tailored to family caregivers remain limited. AI-powered conversational agents offer a promising approach for delivering on-demand, personalized mental health support, yet development and evaluation frameworks for this population are lacking. Objective: This paper describes the iterative development and formative evaluation of COCO (Caring of Caregivers Online), a conversational agent designed for family caregivers of children with chronic health conditions. COCO integrates problem-solving therapy (PST) and motivational interviewing (MI) within a human-in-the-loop development framework that progressed from rule-based interactions to a large language model (LLM)–powered conversational agent. Methods: COCO was developed across four phases: (1) caregiver persona and dialogue development based on PST and MI; (2) usability testing of a low-fidelity prototype with standardized patients in a single session of PST; (3) usability testing of a high-fidelity prototype with caregivers in a single session of PST (n=38); (4) integration of an LLM into COCO. The Wizard-of-Oz method was used across phases 2 and 3 to collect naturalistic dialogues and refine COCO’s conversational design. In phase 3, usability of COCO was assessed using the System Usability Scale (SUS). Caregiver emotions were measured before and after the session using 6 subscales of the PANAS-X. In phase 4, GPT-4 was integrated into COCO with few-shot learning and evaluated by research team members using the caregiver personas. Descriptive statistics were used to summarize quantitative measures. The MI principles and techniques used by COCO across the 4 phases were coded using the . Results: In phase 1, 4 gold-standard dialogues were developed using caregiver personas. In phase 2, standardized patients described COCO as validating and identified its problem-solving and on-demand support as helpful for caregivers. In phase 3, COCO-Wizard-of-Oz achieved a mean SUS score of 75.6% (SD 12.9%), reflecting acceptable usability. Participants demonstrated significant improvement in negative affect, sadness, guilt, and fatigue following PST sessions (

This author reply responds to a Letter to the Editor commenting on our systematic review and meta‑analysis evaluating radiomics‑based artificial intelligence for predicting pathological response following neoadjuvant immunochemotherapy in non‑small‑cell lung cancer. We clarify several methodological points raised in the comment, including patient versus assessment counts in a cited study, cross‑study versus within‑patient comparisons of diagnostic metrics, and the sensitivity‑specificity trade‑off between artificial‑intelligence models and conventional response criteria (RECIST 1.1, PERCIST). We acknowledge two textual errors in the original discussion and confirm they do not affect primary pooled analyses. We further elaborate on eligibility constraints, heterogeneity across prediction time points, definitions of pathological complete response, and reporting standards such as DECIDE‑AI. Our core conclusion remains unchanged: radiomics‑based artificial intelligence shows promising predictive performance with a potential sensitivity advantage over RECIST 1.1, though definitive evidence requires prospective same‑patient, same‑time‑point validation studies.

This letter examines the interpretation and clinical translation of a meta-analysis of radiomics-based artificial intelligence (AI) for predicting pathological response after neoadjuvant immunochemotherapy in resectable non–small cell lung cancer. We highlight that the conventional-response comparator combines PERCIST and RECIST 1.1 assessments from the same 36-patient cohort, whereas the AI estimates arise from unmatched cohorts; consequently, the reported denominator and Z tests do not establish comparative superiority. We further consider how restriction to patients who reached resection and variation in imaging timepoints narrow the clinical estimand and limit inference about earlier treatment redirection. Clarification of the RECIST and error-direction examples is also warranted. We propose same-patient comparisons in treatment-initiation cohorts using fixed imaging times, locked thresholds, explicit primary-tumor and nodal labels, and calibration and net-benefit analyses at prespecified clinical thresholds.

Background: Cardiometabolic diseases (CMDs) are a major health concern worldwide, with people living in socioeconomically disadvantaged areas being disproportionately affected. Knowing one’s risk status for developing a CMD can help individuals in delaying or preventing the disease. Innovative tools and strategies such as the use of AI-based technology are needed to improve inclusivity, cost-effectiveness, and sustainability of screening processes. Objective: This study aimed to assess users’ perceptions of the usability and acceptance of the social robot “Furhat” for cardiometabolic risk assessment within a socioeconomically disadvantaged community in Uppsala, Sweden. Methods: In this qualitative study, 13 semistructured interviews were conducted with participants living in socially disadvantaged areas within Uppsala, after they completed a CMD risk assessment delivered by the social robot. The study was conducted in one of the socioeconomically disadvantaged areas in Uppsala from October 19 to 26, 2023. Participants were purposefully sampled at different community events via phone or on the street to achieve the desired variation regarding personal characteristics such as age, gender, or area of living. The used interview guide for this study consisted of questions regarding the demographics, participants’ experiences with the robot screening, and suggestions for improvements. A framework analysis approach was applied to the data. Results: Two themes were developed. The first theme includes participants’ perceptions of who would use the robot and why, if it was installed in the community. Participants perceived older individuals, immigrants, and, in some instances, women to be less interested in the robot screening or as having a harder time participating in it. However, there were between-group variations, especially among men and women, with participants often discounting opinions ascribed to their group by others. The second theme captures participants’ actual experiences of the interaction with the social robot once they have started the interaction. Although participants enjoyed the interaction with the robot, perceived it as nonjudgmental, and would recommend it to others, most would prefer conducting the screening with a human. Two major contributors for this preference were language barriers experienced during the robot-participant interaction, as well as their wish for more interactive, emotionally responsive, and communicative features of the robot. Conclusions: This study provides insights into users’ perceptions of the usability and acceptance of a social robot for risk assessment in a community context, which can help inform the further development of this technology for health screening in a community context. In the future, the robot could be further developed to conduct risk assessments in multiple languages and offer a broader service to users, such as providing appropriate information related to healthy and active living.

AI is increasingly embedded in the institutions and environments that shape health. Yet current frameworks for understanding its implications for health equity remain underdeveloped. The social determinants of health tradition provides a strong foundation, and recent work on digital determinants of health has begun to address the health implications of digital transformation. However, AI warrants distinct conceptual attention because of its triple role: it operates simultaneously as a determinant of health in its own right, as a mediator and moderator of existing determinants, and as an amplifier of advantage and disadvantage over time. This viewpoint proposes a redistribution-translation-accumulation framework for analyzing how AI may contribute to the reproduction of health inequity. The framework comprises 2 analytically distinct mechanisms and 1 cross-cutting temporal dynamic. Redistribution captures how AI reshapes the distribution of health-relevant resources and opportunities, including education, employment, and income, while AI itself becomes an unequally distributed determinant. Translation describes how AI changes the pathways through which social positions are converted into health outcomes. Proxy-based decision rules can formalize historical inequities, diagnostic algorithms may perform unevenly across populations due to unrepresentative training data, and AI-mediated information environments can alter institutional responsiveness. Accumulation is conceptualized not as a third parallel mechanism but as a temporal amplifier operating on both mechanisms: AI-driven feedback loops and institutional embedding can concentrate advantage and disadvantage over time, often without users’ awareness. The framework is offered as a hypothesis-generating heuristic and is directionally neutral: under specifiable design, deployment, and governance conditions, the same mechanisms can narrow rather than widen health gaps in high-income and low- and middle-income settings alike. The framework has direct implications for governance. Current approaches such as the EU AI Act’s Fundamental Rights Impact Assessment (FRIA) and Canada’s Algorithmic Impact Assessment (AIA) advance AI accountability but assess systems largely before or at deployment and do not systematically track distributional health consequences. Building on this framework, I propose a distributional impact assessment as a complementary tool for equity-oriented AI governance. Structured around the 3 RTA dimensions, it asks whether an AI system alters the distribution of health-relevant resources across groups (redistribution), changes how social positions are converted into health (translation), and risks concentrating disadvantage over time through feedback and institutional embedding (accumulation). It is operationalized with candidate indicators, data sources, responsible actors, and reassessment triggers and is illustrated through a retrospective worked example of a biased care-management algorithm.

Background: Thyroid fine-needle aspiration biopsy (FNAB) is a commonly used diagnostic procedure in patients with suspected thyroid cancer; however, it may induce pain, anxiety, and fear during the procedure. Objective: This open-label randomized controlled trial aimed to evaluate the effect of virtual reality (VR) on pain as the primary outcome and anxiety and fear of pain as secondary outcomes in patients undergoing thyroid FNAB. Methods: The study was conducted between January 19, 2025, and April 30, 2025, at Gaziantep City Hospital, Türkiye. A total of 100 patients with suspected thyroid nodules were randomly assigned to either a VR intervention group (n=50) or a control group (n=50). Data were collected using a patient information form, the visual analog scale (VAS), the Beck Anxiety Inventory (BAI), and the Fear of Pain Questionnaire-III (FPQ-III). Results: After adjustment for baseline pain, previous thyroid mass diagnosis, and voice tone changes, the VR group had statistically significantly lower postintervention pain scores than the control group (adjusted mean 3.627 vs 4.493; F_1,95=4.021; P=.048; partial η²=0.041). However, the unadjusted between-group comparison for pain was not statistically significant (P=.12), and the unadjusted effect size was small, with a 95% CI that crossed 0 (Cohen d=−0.33, 95% CI −0.72 to 0.07). No statistically significant adjusted between-group differences were observed for anxiety (P=.48) or fear of pain (P=.07). Unadjusted standardized between-group effect sizes were also small for anxiety (d=−0.17) and fear of pain (d=−0.11). Conclusions: The adjusted analysis suggested a small reduction in procedural pain with VR; however, the between-group difference was not statistically significant in the unadjusted analysis and reached statistical significance only after adjustment for baseline pain and 2 nonprespecified covariates selected on the basis of observed baseline imbalance. Moreover, the observed adjusted effect (f=0.207) was smaller than the minimum effect size the trial was powered to detect (f=0.283). No statistically significant adjusted between-group effects were found for anxiety or fear of pain. Therefore, the potential analgesic effect of VR should be interpreted cautiously and confirmed in larger, adequately powered trials. Trial Registration: ClinicalTrials.gov NCT06792929; https://clinicaltrials.gov/study/NCT06792929

npj Digital Medicine, Published online: 06 October 2026; doi:10.1038/s41746-026-03367-x

Virtual reality for cognitive and neuropsychiatric outcomes in mild cognitive impairment and dementia: a network meta-analysis

npj Digital Medicine, Published online: 06 October 2026; doi:10.1038/s41746-026-03273-2

Calibrated AI approach to pharmacovigilance using FAERS

npj Digital Medicine, Published online: 06 October 2026; doi:10.1038/s41746-026-03359-x

LLM-derived narrative metrics in chronic pain: convergent validity and novel cognitive constructs

npj Digital Medicine, Published online: 06 October 2026; doi:10.1038/s41746-026-03345-3

A legal and ethical architecture for a surgical video data registry in England

npj Digital Medicine, Published online: 06 October 2026; doi:10.1038/s41746-026-03268-z

Machine learning classification of digital exclusion in ageing populations across five international cohorts

npj Digital Medicine, Published online: 06 October 2026; doi:10.1038/s41746-026-03318-6

SPIRIT-CONSORT-ELM: element-level annotated dataset and large language model approach for assessing randomized controlled trial reporting

Background: Online health communities (OHCs) provide vital peer support and health information to individuals managing chronic conditions. However, sustained user engagement remains challenging, with many users reducing activity over time despite the ongoing benefits these platforms offer. While some individuals may improve or become more knowledgeable, sustained engagement remains critical because ongoing participation fosters trust, peer support, and continuous access to evolving health information that may persist beyond initial recovery or learning. Hence, it is important to consider how community needs evolve as users’ health needs and information-seeking behaviors change to inform how we support users through different stages of their health and participation journeys. Objective: The aim of the study is to examine user engagement in a large OHC, identifying perceived stage-based behaviors, motivation, barriers, and design opportunities that could facilitate progression between stages and enhance long-term participation. Methods: We conducted semistructured interviews with 19 members of the American Heart Association Support Network Community. Participants were patients or survivors managing various cardiovascular diseases. Using narrative thematic analysis, we examined users’ perceived engagement motivation, behavior, challenges, and design opportunities across different stages of their community involvement. Results: This study highlighted 4 distinct engagement stages: discovery (crisis-driven initial engagement), exploration (navigation and orientation), commitment (active engagement and information management), and integration (sustained engagement and mentorship). Key barriers included information architecture complexity, concerns about misinformation, limited support for role transitions, and decreased participation as health management improved. Participants identified opportunities through which OHCs could increase long-term engagement, including adaptive recommendation systems, health information literacy programs, structured role transition support, and alternative engagement modalities, such as synchronous interactions and health tracking tools. Conclusions: User engagement in OHCs is dynamic and evolves with changes in health status, knowledge, and personal circumstances. Supporting sustained engagement requires stage-appropriate interventions, including personalized content delivery, health information literacy education, structured pathways for role transitions, and diversified engagement options. These findings provide actionable insights for designing OHCs that better support users throughout their health journey.

npj Digital Medicine, Published online: 05 October 2026; doi:10.1038/s41746-026-03340-8

Telemedicine in surgical and anesthetic care in urban and rural settings across time: a scoping review

npj Digital Medicine, Published online: 05 October 2026; doi:10.1038/s41746-026-03232-x

Standardized pre-consultation by a large language model agent vs ophthalmology residents: a randomized clinical trial

Oncogene, Published online: 04 October 2026; doi:10.1038/s41388-026-04002-5

KDM4A drives TGCT metastasis by inducing focal adhesion disassembly via STAT1-mediated CCL3 transcriptional activation

Oncogene, Published online: 02 October 2026; doi:10.1038/s41388-026-03990-8

Colorectal cancer is often treated with oxaliplatin, but many tumors become less responsive over time, making treatment harder. This study aimed to understand one way cancer cells escape oxaliplatin and to identify a possible way to restore drug response. The researchers studied colorectal cancer cells, drug-resistant cells, mouse tumor models, and tumor-like structures grown from patient samples. They found that oxaliplatin stress lowers a mitochondrial protein called TFAM. This allows mitochondrial DNA to leak into the cell fluid and switch on a survival signal involving TBK1. This signal helps cancer cells clear damaged cell parts and survive treatment. Blocking TBK1 reduced this protective response and made resistant tumors more sensitive to oxaliplatin. These findings suggest that combining oxaliplatin with a TBK1-blocking treatment may help overcome drug resistance in colorectal cancer in the future.