Healthcare

Introduction: Managing gestational weight gain (GWG) is crucial for the health of mothers and their children. Mobile applications (apps) specifically designed for pregnancy are emerging as modalities to deliver accessible lifestyle intervention at a low-cost. However, current studies are varied in results and suffer from heterogeneity. Thus, we conducted this systematic review and meta-analysis to summarize the efficacy of mobile apps in managing GWG and investigate variables that may contribute to heterogeneity. Methodology: Seven databases were systematically searched up to 9 November, 2024. Only randomized controlled trials (RCTs) were included. Outcomes were excessive GWG and inadequate GWG according to the 2009 Institute of Medicine (IOM) guideline. Quality appraisal was performed using the Cochrane Risk of Bias 2 (RoB 2) tool. Random-effect model meta-analysis was conducted using odds ratio (OR) as the summary measure alongside their 95% confidence intervals (CI). Results and Discussion: Fifteen RCTs were included. Mobile apps led to a significant overall decrease in excessive GWG (OR: 0.71; 95% CI: 0.54 to 0.95; p-value: 0.02; I2: 60%). Subgroup analysis showed that social media apps, self-monitoring functionalities, and overweight/obese patients are associated with a significant reduction in excessive GWG. However, there was significant evidence of small-study bias in the analysis. Moreover, mobile apps also significantly increased inadequate GWG (OR: 1.51; 95% CI: 1.04 to 2.21; I2: 0%). Meta-regression did not reveal any significant finding. Conclusion: In conclusion, mobile app interventions are shown to be effective in preventing excessive GWG, particularly social media apps and those with self-monitoring functionalities. However, the reduction in excessive GWG may only be seen in overweight and obese patients and more studies are needed to ascertain this finding. Lastly, mobile apps are associated with an increased risk of inadequate GWG and strategies to combat inadequate GWG are needed.

Background: Occupational therapists can improve stroke survivors hand and arm movement and participation in daily activities through action observation (AO). AO involves watching another persons hand or arm complete a movement or task. While research generally supports the use of AO with stroke survivors, there are limited AO videos are available to occupational therapists which makes applying AO challenging. Objective: The purpose of this work is to develop structured and widely accessible tool to support access to AO for stroke survivors, occupational therapists, and researchers. Methods: To develop an AO video library for stroke rehabilitation, functional and non-functional upper limb task deficits were first identified through clinical observations and clinician interviews to establish a prioritized list of daily activities. In collaboration with media production specialists, healthy adult volunteers were recruited and filmed performing these tasks from both first- and third-person perspectives. The recorded videos were then systematically edited, enhanced with instructional title slides, and distributed via a public YouTube channel for clinical application and a categorized digital repository for research purposes. Results: Initial assessments revealed a complete lack of familiarity, awareness, and utilization of AO resources among local occupational therapists, despite high perceived clinical utility. To address this gap, a final library of 150 tasks was established, resulting in the production of 419 finalized, standardized videos featuring six healthy volunteers. For clinical application, these videos were hosted on a free, public YouTube channel organized into 18 functional playlists, while a parallel set was structured into distinct movement categories for research repository storage. Conclusion: By providing a structured and highly accessible tool, this repository enables clinicians, researchers, and caregivers to readily implement evidence-based action observation interventions in both clinical and home settings.

Objective: Hispanic/Latinx populations in the U.S. experience higher rates of chronic disease linked to physical inactivity, yet digital health interventions remain largely inaccessible to more than 16 million Hispanic/Latinx adults with limited English proficiency. While large language models (LLMs) offer scalable personalization, their use in non-English behavioral coaching is unexplored. This study introduces MHC-Coach-ES, a Spanish-language LLM fine-tuned on the Transtheoretical Model (TTM) of behavior change. Materials and Methods: We fine-tuned Llama 3-70B-Instruct using a two-stage pipeline. First, the model was adapted to Spanish health and motivational language using a 2.21-million-token corpus. Second, it was instruction-tuned on 3,268 translated human written messages to align the model with the Transtheoretical Model (TTM) of Behavioral Change. We compared MHC-Coach-ES with Llama 3-70B-Instruct and translated human-expert messages using a forced-choice preference survey (N = 77) and blinded expert review (N = 2). Results: Spanish-speaking participants significantly preferred MHC-Coach-ES messages over translated human-expert messages (81% preference, P<0.001). Linguistic analysis showed that MHC-Coach-ES produced more temporally anchored messages than the base model (65% vs. 20%), while maintaining readability. In blinded evaluation, clinical experts rated MHC-Coach-ES higher for alignment with Transtheoretical Model stages than human-expert messages (4.83 vs. 4.38 out of 5). The base model also outperformed translated expert messages across preference and expert ratings. Conclusions: Generative AI can operationalize behavioral science frameworks in Spanish, offering a scalable approach to reducing health disparities. The strong performance of both MHC-Coach-ES and the base model highlights the promise of generative and personalized approaches over translation-based localization for theory-driven behavioral interventions.

BackgroundGenerative artificial intelligence (GenAI) chatbots have shown utility in assisting with various research tasks. Traditional, complementary, and integrative medicine (TCIM) is a patient-centric approach that emphasizes holistic well-being. The integration of TCIM and GenAI presents numerous key opportunities. However, TCIM researchers attitudes toward GenAI tools remain less understood. This large-scale, international cross-sectional survey aimed to elucidate the attitudes and perceptions of TCIM researchers regarding the use of GenAI chatbots in the scientific process. MethodsA search strategy in Ovid MEDLINE identified corresponding authors who were TCIM researchers. Eligible authors were invited to complete an anonymous online survey administered via SurveyMonkey. The survey included questions on socio-demographic characteristics, familiarity with GenAI chatbots, and perceived benefits and challenges of using GenAI chatbots. Results were analysed using descriptive statistics and thematic content analysis. ResultsThe survey received 716 responses. Most respondents reported familiarity with GenAI chatbots (58.08%) and viewed them as very important to the future of scientific research (54.37%). The most acknowledged benefits included workload reduction (74.07%) and increased efficiency in data analysis/experimentation (71.14%). The most frequently reported challenges involved bias, errors, and limitations. More than half of the respondents (57.02%) expressed a need for training to use GenAI chatbots in the scientific process, alongside an interest in receiving training (72.07%). However, 43.67% indicated that their institutions did not offer these programs. ConclusionBy developing a deeper understanding of TCIM researchers perspectives, future AI applications in this field can be more informed, and guide future policies and collaboration among researchers.

Background & AimsAccurate assessment of clinical malnutrition using anthropometric and functional indicators could improve the care of elderly trauma patients in intensive care units (ICUs). This study aimed to develop an AI-driven malnutrition assessment toolbox based on a minimal set of clinically feasible indicators. MethodsMultiple machine learning models, including logistic regression, support vector machines, k-nearest neighbors, decision trees, random forests, XGBoost, and neural-network-based ensemble models, were developed using different indicator configurations from a clinically collected patient dataset. Models were trained using baseline and longitudinal measurements to predict malnutrition risk. SHAP analysis was used to interpret the importance of selected indicators. ResultsBaseline (Day 1) data alone did not provide a reliable prediction, whereas longitudinal measurements substantially improved performance. Models based on a minimal indicator set, including bilateral mid-upper arm circumference, calf circumference, and key static variables, outperformed models using the full indicator set. Tree-based methods consistently outperformed linear and distance-based models, with the three-time-point XGBoost achieving the best individual performance. Neural-network-based ensemble models further improved predictive stability. The best overall performance was achieved by the ensemble model using the minimal indicator set from Day 1 and Day 3. SHAP analysis confirmed the importance of the selected indicators. ConclusionsThis AI-driven toolbox provides an efficient and clinically feasible approach for early malnutrition assessment in elderly trauma patients in the ICU. Its strong performance with a minimal indicator set supports its potential for integration into clinical workflows and future digital twin systems for intelligent nutritional management.

Background. People with lived/living experience of health conditions, as well as caregivers, are increasingly engaged in research. This study aimed to develop and pilot test a new tool measuring the impact of lived/living experience engagement on the research. The measure is called the Measure of Engagement Tool for Research and lived Experience (METRE). Method. We conducted a qualitative descriptive study among 28 people with lived/living experience and caregivers and 12 academic researchers to understand the impacts of engagement. Using the findings, we drafted the METRE. We pilot tested the METRE among 13 people with lived/living experience and caregivers and 10 academic researchers. Insights were used to refine the scale. Results. Qualitatively, participants identified multiple domains of impact of engagement on research, which guided scale development. Pilot testing of the draft METRE revealed it being straightforward to complete, providing a thorough evaluation of the impact of engagement. However, some areas of improvement were recommended. The draft items showed acceptable preliminary performance. Conclusions. An assessment tool is now available to assess the impact of lived/living experience engagement on the research. Additional research is required to evaluate its psychometric properties. Tools to evaluate the impact of engagement on research will help advance the science of engagement and support engaged research teams in their work.

BackgroundThe COVID-19 pandemic significantly disrupted healthcare services globally, particularly in low-resource settings. This study explores the impact of the pandemic on physiotherapy services in Nepal. MethodsA cross-sectional study was conducted. Qualitative data were collected through semi-structured interviews with 12 physiotherapists, while quantitative data were gathered from an onsite survey of 29 health facilities at six different districts of Province III of Nepal. Inductive thematic analysis approach was used to analyze the qualitative data, and descriptive statistics were used for the closed ended questions. ResultsThe findings were categorized into sub-themes under two major themes: i) Pandemic effect on physiotherapy services and patient care and ii) Adaptation, innovation and collaboration. The study revealed a significant disruption in physiotherapy services with a notable decline in patient flow and service availability. Most patients, especially those with disabilities and post-operative needs, experienced worsening conditions due to limited access to care. There was an increased recognition of the role of physiotherapy in acute respiratory care and post-COVID-19 recovery. Tele-rehabilitation was explored as an alternative care method but faced challenges in implementation. More than half (62.07%) of the centers reported uninterrupted physiotherapy services, whereas almost one third (31.03%) experienced service suspension. Most centers (89.7%) had personal protective equipment available, and majority (86.2%) of the physiotherapists worked in multidisciplinary team: fever clinics, triage, emergency care, respiratory physical therapy, and nursing and administrative support were among the expanded roles. Several centers (37.9%) used virtual care with telephone consultation serving as the primary modality. Virtual service was mostly absent in centers where in-person services persisted. ConclusionThe COVID-19 pandemic significantly impacted physiotherapy services in Nepal, leading to service disruptions and compromised patient care. It highlighted the need to further incorporate physiotherapy into the healthcare system and enhance rehabilitation services to improve continued patient care.

Abstract Objective: Frailty is an important concern in old age. Inflammation can cause frailty. Anti-inflammatory food supplements can play a role in slowing down frailty processes and consequences. This study explored the views of people (aged 50-89 years) on the need to develop a frailty supplement, preferences for its form and how older people could be encouraged to use such a supplement. Design: We conducted semi-structured qualitative interviews and used a framework method to analyse the data. Participants: 30 participants from a city in the UK. Setting: These participants were recruited from social housing, care homes, foodbanks and the wider population. Participants were from diverse ethnic, gender and age backgrounds. Results: Participants identified a strong need for the development of a food-based supplement for frailty. They expressed excitement for the supplement and viewed it as something which they would be happy to integrate in their daily food routine. In terms of preferences, our participants wanted to have multiple options, however, a biscuit-based supplement was preferred by most. The participants preferences were mainly based on taste of the supplement, its effectiveness, convenience in use and affordability. Muslim participants in the sample said they would be happy to use this supplement if it was developed using Halal ingredients. In terms of creating awareness and encouraging people to use the proposed supplement, participants suggested a variety of marketing methods. These included: word of mouth, face to face sessions with older adults, social media, especially YouTube and advertising on TV. Conclusion: The participants were generally open to the idea of a food-based supplement and felt that it could easily fit with their existing food practices and lifestyles. Keywords: older adults, frailty, food supplement, co-creation, healthy ageing

The integration of artificial intelligence (AI) in cardiology requires healthcare worker acceptance for successful implementation. Understanding attitudes and educational needs is crucial for developing effective training programs. A cross-sectional survey was conducted among 408 healthcare workers treating cardiac diseases in Riyadh, Saudi Arabia. We assessed AI acceptance, knowledge levels, and training preferences using validated scales. Statistical analyses included descriptive statistics, chi-square tests, correlation analysis, reliability testing, and multiple logistic regression. Of 408 participants, 407 provided complete responses. The sample comprised predominantly young (87.0% aged [&le;]30), female (75.7%) medical residents (89.9%) with limited AI experience (86.7% never used AI clinically). Internal consistency was excellent (Cronbachs = 0.892). Moderate acceptance was observed: 49.9% were aware of AI applications in cardiology, 46.7% were willing to learn, and 42.8% were willing to use AI clinically. However, 49.1% acknowledged lacking sufficient AI knowledge. Logistic regression identified willingness to learn (OR = 3.24, 95% CI: 2.15-4.89) and training interest (OR = 2.87, 95% CI: 1.94-4.25) as the strongest predictors of AI acceptance. The model explained 68.4% of variance (Nagelkerke R{superscript 2} = 0.684) with an AUC of 0.847. Medical residents demonstrate moderate AI acceptance but significant knowledge gaps. Educational interventions--particularly hands-on learning and institutional training programs--are the strongest drivers of AI readiness, surpassing demographic predictors. Integrating AI literacy systematically into medical curricula is essential for successful AI adoption in cardiovascular care. Author summaryHealthcare workers worldwide are increasingly encountering artificial intelligence (AI) tools in clinical settings, yet their readiness to adopt these technologies--particularly in specialized fields like cardiology--remains poorly understood, especially in rapidly developing healthcare systems. In this study, we surveyed 407 healthcare workers in Riyadh, Saudi Arabia, to understand their current attitudes, knowledge gaps, and learning preferences regarding AI in cardiac diagnosis. Our findings reveal that while most participants hold cautious optimism about AI, nearly half acknowledge lacking the knowledge needed to use it confidently. Crucially, we found that educational factors--specifically willingness to learn and interest in institutional training--were far stronger predictors of AI acceptance than demographic characteristics such as age or gender. This means that AI readiness is not a fixed trait determined by who someone is, but a teachable and trainable capacity. These results carry direct implications for medical educators and policymakers: structured, hands-on AI training integrated throughout medical curricula can meaningfully accelerate adoption of beneficial technologies in cardiovascular care and beyond.

The postpartum period involves profound physical, emotional, and social changes, yet many women report fragmented, infant-centered care that leaves their own needs insufficiently addressed. Digital health tools, including mobile apps, wearables, telehealth, and online resources, are increasingly used by postpartum women to seek information, support, and reassurance; however, little is known about how women experience these tools in their everyday lives. This qualitative study employed thematic analysis to explore the perspectives of postpartum women on digital health. Postpartum women ([&le;]12 months after birth) living in Alberta, Canada, were recruited through maternity clinics and targeted social media advertisements. Four virtual focus groups (4-6 participants in each; 18 participants overall) were conducted via Zoom using a semi-structured guide on postpartum healthcare experiences, use of digital tools (apps, wearables, telehealth, AI), and perceived barriers and facilitators to adoption. Sessions were audio-recorded, transcribed verbatim, and coded by multiple researchers. Thematic analysis identified 32 codes, organized into 12 subthemes and four overarching themes: navigating postpartum support networks; empowerment through digital health tools; conditions for acceptable digital health design; and when technology feels like a burden. Women appreciated multiple sources of support from midwives, public health nurses, peers, and online communities, but described care that quickly became infant-focused, leaving their own recovery and mental health under-addressed, particularly in rural settings. Digital tools helped mothers structure infant and self-care, track symptoms, and prepare for appointments, yet also created new forms of burden, including information overload, usability challenges, privacy concerns, and feelings of surveillance or pressure to perform. Participants emphasized personalization (flexible notifications, mother-focused content), embedded mental health support, integration with trusted providers, and co-designed, credible platforms endorsed by Canadian health systems. Overall, to be acceptable and effective, tools must center mothers needs and be embedded within a broader ecosystem of responsive, continuous care. Author summaryBecoming a parent is a major life change, and many women feel that support from the health system drops off once the baby is born. At the same time, new mothers are increasingly turning to mobile phone apps, wearable devices, online groups, and video visits to answer questions, track health, and feel less alone. We wanted to understand women lived experience with these digital tools after giving birth: what feels helpful, what feels burdensome, and what they would want in an ideal tool. Our research team, consisting of three PhD students, held four online focus group discussions (4-6 participants per group; 18 participants overall) with women in Alberta, Canada, who had given birth within the past year. They described digital tools as both empowering and exhausting. Apps and wearables helped them track feeding, sleep, and symptoms, organize daily life, and come better prepared for medical appointments. At the same time, constant tracking, frequent notifications, and unclear data practices could feel overwhelming, guilt-inducing, or intrusive. This study is an important first step in a larger co-design work. By listening closely to mothers stories, we gathered practical ideas about what a supportive postpartum app should (and should not) do. In future phases, we plan to work directly with postpartum women and frontline clinicians to turn these ideas into a user-friendly, trustworthy digital tool that supports both mothers and babies health.

Background: The Dementia Supporter Initiative is a national public education program in Japan that aims to foster positive attitudes and appropriate understanding of dementia to support people with Alzheimer's disease and related dementia in the community. However, its influence on the community as a whole remains unclear. Objective: This study examined the relationship between dementia supporter training and residents' attitudes and recognition related to dementia at the municipal level. Methods: This ecological cross-sectional study linked municipal-level data from the Japan Gerontological Evaluation Study 2022 wave with publicly available information on the number of dementia supporters. Residents' beliefs and attitudes toward dementia and recognition of dementia consultation services were assessed by mail questionnaires and aggregated at municipal level. The proportion of dementia supporters in each municipality was calculated as of September 2022. Results: Data from 69 municipalities were analyzed. The mean proportion of dementia supporters was 13.47% (2.62-44.85). A higher proportion of dementia supporters was positively correlated with community support-seeking for a family member with dementia (r = 0.328) and recognition of dementia consultation services (r = 0.501). Regression analysis adjusted for municipal covariates also showed their positive associations (per 10-percentage-point increase: coef. = 1.44, p = 0.047; coef. = 3.12, p < 0.001, respectively). No associations were observed with residents' positive attitudes and appropriate understandings of dementia. Conclusions: Wider dissemination of dementia supporters may contribute to better recognition of community support resources, but may be insufficient to influence broader public attitudes and understanding of dementia at the community level.

Global healthcare is targeting patient-centred care, as it leads to better health outcomes and higher level of patient satisfaction. Patient-centred communication, is an important part of patient-centred care because it focuses on involving patients in their care. Recent surveys both nationally and globally have shown that patients are not involved enough in their own healthcare decisions. This problem is especially common among the elderly with chronic conditions. This study aimed to describe patient-healthcare professional interactions, expectations, and satisfaction in physiotherapy within an understudied context, thereby providing important, specific data on ICE dynamics and satisfaction in the specific setting. Cross-sectional study of participants in scheduled consultations was conducted. Two government physiotherapy centres, seven private physiotherapy centres and two trust centres with physiotherapy facilities in Gujarat, India. 232 patients (from various public and private physiotherapy clinics) participated in the study. Patients' ideas, concerns, expectations (ICE) and satisfaction were explored. Almost 88% of patients reported their thoughts and explanations about their symptoms during the consultation. Most patients described not having any concerns about the diagnosis/treatment, and more than two-third of patients consulting PTs expected explanation for their symptoms. Almost 90% patients were satisfied with the consultation. The study revealed that while most patients conveyed their thoughts during consultations, very few expressed their concerns. Overall, patients were satisfied with their consultations.

While vaccination conflicts have become apparent, physicians attitudes toward those with differing views remain unclear. Through an online survey of 492 physicians and 5,252 members of the general public in Japan in February 2026, we investigated attitudes toward four vaccines (influenza, measles, HPV, and COVID-19). Intergroup bias was assessed as ingroup minus outgroup attitudes using a feeling thermometer. Multilevel regression examined associations with agreement group and physician status. Intergroup bias was significantly positive in both agreement and disagreement groups across all vaccine types, and was higher in the agreement group. Physicians exhibited higher intergroup bias than the general public. These findings indicate that vaccination conflict is bidirectional: physicians, often viewed as targets of hostility from vaccine-hesitant individuals, themselves exhibit greater intergroup bias toward those with opposing views. Interventions to raise physicians awareness of their own bias, alongside communication strategies for vaccine-hesitant individuals, are needed.

Background: Clinical documentation and information retrieval consume over half of physicians working hours, contributing to cognitive overload and burnout. While artificial intelligence offers a potential solution, concerns over hallucinations and source reliability have limited adoption at the point of care. Objective: To evaluate clinician-reported time savings, decision-making support, and satisfaction with DR. INFO, an agentic AI clinical assistant, in routine clinical practice. Methods: In this prospective, single-arm pilot study, 29 clinicians across multiple specialties in Portuguese healthcare institutions used DR. INFO v1.0 over five working days within a two-week period. Outcomes were assessed via daily Likert-scale evaluations and a final Net Promoter Score. Non-parametric methods were used throughout. Results: Clinicians reported high perceived time saving (mean 4.27/5; 95% CI: 3.97-4.57) and decision support (4.16/5; 95% CI: 3.86-4.45), with ratings stable across all study days and no evidence of attrition bias. The NPS was 81.2, with no detractors. Conclusions: Clinicians across specialties and career stages reported sustained satisfaction with DR. INFO for both time efficiency and clinical decision support. Validation in larger, controlled studies with objective outcome measures is warranted. Keywords: Medical AI assistant, LLMs in healthcare, Agentic AI, Clinical decision support, Point of care AI

BackgroundHealthcare providers in low- and middle-income countries (LMICs) are increasingly relying on Artificial Intelligence (AI) tools, yet most available AI assistants are general-purpose systems not designed for the specific clinical, epidemiological, and resource contexts of these settings. There is no evidence, from physicians assessments, on whether clinical reasoning support from purpose-built, context-specific and retrieval-augmented AI tools can outperform general-purpose AI agents. MethodsWe conducted a prospective multi-site validation study enrolling 37 physicians across India and Bangladesh. Each physician evaluated two AI tools (a) VITA (Validated Intelligence for Treatment and Assessment), a purpose-built (context-specific and retrieval-augmented) clinical reasoning AI assistant trained on India-specific guidelines, antimicrobial resistance patterns, and formulary constraints, and (b) ChatGPT Plus (version 5.2), a leading general-purpose AI assistant on six hypothetical clinical case vignettes (three predefined, three physician-selected). Evaluations were scored across six dimensions (differential diagnosis, clinical workup, treatment recommendation, dosing, clinical decision-making, and evidence quality) on a 1-5 Likert scale, yielding 444 observations. Analyses included paired t-tests, Wilcoxon signed-rank tests, and multivariate regressions with robust standard errors. ResultsVITA scored significantly higher than ChatGPT across all six evaluation dimensions. The mean composite score (sum of all dimensions, maximum = 30) was 25.4 for VITA versus 22.3 for ChatGPT (difference = +3.1 points, t = 8.31, p < 0.001). The largest advantage was in evidence quality (VITA: 4.46 vs. ChatGPT: 3.14, a 42% relative gap). VITAs advantage was consistent across both predefined and doctor-defined hypothetical cases and was robust to controls for physician demographics, case type, and evaluation order in multivariate regression (coefficient = +3.08, p < 0.001). ConclusionsIn this first systematic head-to-head physician evaluation of a purpose-built clinical reasoning AI assistant versus general-purpose AI in an LMIC setting, physicians consistently rated the context-specific tool as superior. These findings suggest that contextual relevance--including local guidelines, formulary constraints, and resistance patterns--matters for clinical AI adoption and quality in resource-limited settings.

Introduction Patients hospitalized with heart failure who do not speak English as their primary language face communication barriers, however the impact on documented History of Present Illness (HPI) and Review of Systems (ROS) has not been reported. Methods This retrospective cohort study was based on MIMIC-IV, an anonymized clinical database. Adult patients admitted to general medicine or cardiology services with heart failure (by DRG) were identified. Multivariable linear regression was used to assess for an association between language (English vs. non-English) and word counts for HPI+ROS and HPI word counts. Qualitative differences in texts were also analyzed using Claude Opus 4.6. Results In a cohort of 552 patients, non-English language (N = 81) was associated with a shorter HPI+ROS (coef. -33.387, 95% CI [-62.076, -4.697], p = 0.023) controlling for age (coef. -1.023, 95% CI [-1.817, -0.230], p = 0.012) and Elixhauser score (coef. 10.391, 95% CI [7.078, 13.705], p<0.001). Similar associations were found for HPI alone. Qualitative differences included less discussion of symptoms and timing of onset. Discussion HPI+ROS and HPI were more abbreviated when the primary documented language was not English. This has important implications for equitable care and the development of emerging translation and documentation technologies.

Ambient AI documentation systems rely on automatic speech recognition to transcribe patient-provider conversations before generating clinical notes. However, little empirical evidence exists on how these systems perform in mixed-language clinical encounters. We conducted a mixed-method heuristic evaluation of an ambient AI documentation tool using 24 reenacted primary care conversations involving Spanish-English and Mandarin-English code-switching. Quantitative analyses measured mixed error rate (MER) and code-switching detection. Overall MER was low, with a median of 4% and less variation in Spanish-English conversations, and 9% in Mandarin-English conversations, but with outliers reaching 67%. The system generally detected language switches reliably, although deletions occurred frequently in Mandarin-English transcripts at switch points. Qualitative analysis revealed transcription errors related to phonetic similarity, automatic language translation, clinical terminology recognition, and language-specific challenges. These findings highlight considerations for improving ambient AI clinical documentation systems to support multilingual providers in delivering care for linguistically diverse populations.

AimsWe aimed to examine public perceptions of sharing various types of health data relevant for AI development, including electronic health records, audio recordings of consultations, medical images, and genetic information, with actors from either the public or the private sectors. MethodsWe analysed data from 38,740 participants of the Health in Central Denmark survey conducted in 2024. Participants were asked whether they would share different types of health data with an AI solution in healthcare. Each participant was randomised to either of two versions of the scenario and question where the AI application was developed in the public or private sector. Descriptive results (proportions and percentages) were weighted to represent the background population of approx. 1 million people in the Central Denmark Region. The association between randomization group (data recipient) and data sharing attitude ("Yes", "No", "Dont know") was analysed using multinomial logistic regression with "Dont know" as reference category. ResultsParticipants were most willing to share medical images (46%), followed by text from patient journals (39%), genetic information (35%), and audio recordings (27%). There were 12-16% higher proportions of willingness to share with public institutions than with private institutions. A high level of uncertainty was observed for all data types (29-36%) regardless of data recipient. Odds ratios ranged from 1.37 to 1.78 for responding "Yes", and from 0.51 to 0.67 for responding "No" to sharing data with public institutions compared to private institutions. ConclusionsPublic acceptance of health data sharing for AI depends on both the perceived sensitivity of the data and the institutional context of use. Strong public governance, transparent safeguards, and clear communication about data use may be important for maintaining trust and enabling responsible development of AI in healthcare.

ObjectiveTo explore the application effect of AI agent-assisted root cause analysis in the management of quality inspection defects in the Central Sterile Supply Department (CSSD) and to systematically compare it with traditional manual analysis methods. MethodsA retrospective case simulation comparative study was conducted. Thirty typical CSSD quality inspection defect cases were selected. Root cause analysis was performed independently by an AI agent-assisted analysis group and a traditional manual analysis group. Using the consensus results of a high-level expert panel as the "gold standard," a quantitative comparison was made across four dimensions: analysis quality, efficiency, practicality, and process experience, employing t-tests and Mann-Whitney U tests. ResultsCompared with the traditional method, the AI-assisted group demonstrated a significantly higher root cause identification accuracy rate (85.6% vs. 72.3%, P<0.001), superior analysis depth (4.4 points vs. 3.6 points, P<0.001), significantly shorter time consumption per case analysis (18.5 minutes vs. 35.2 minutes, P<0.001), and generated more innovative corrective measures (1.8 items/case vs. 0.7 items/case, P<0.001). There was no statistically significant difference between the two groups regarding the feasibility of the proposed measures (4.0 points vs. 4.2 points, P>0.05). ConclusionThe AI agent-assisted root cause analysis method significantly improves the accuracy, depth, and efficiency of analyzing quality inspection defects in the CSSD and facilitates the discovery of more innovative solutions, demonstrating high application value and promotion potential. Implications for Nursing ManagementThis study provides empirical evidence that AI agent technology can be integrated into CSSD quality management to enhance defect analysis efficiency and accuracy. Nursing managers should consider adopting AI-assisted tools to standardize root cause analysis processes, reduce reliance on senior staff experience, and enable faster, data-driven decision-making. The reduced training burden and improved novice performance suggest that AI can help address workforce skill gaps. Future implementation should focus on human-AI collaboration, with managers ensuring adequate training, maintaining human oversight, and periodically updating the knowledge base to reflect local clinical contexts.

ObjectiveTo systematically evaluate the diagnostic performance of large language models (LLMs) in automated medical literature screening and to determine their potential role in supporting evidence synthesis workflows. MethodsA systematic review and meta-analysis was conducted according to PRISMA DTA guidance. PubMed, Web of Science, Embase, the Cochrane Library and Google Scholar were searched from 1 January 2022 to 17 November 2025. Studies assessing LLMs for automated title and abstract screening or full-text eligibility assessment in medical literature were included. Diagnostic accuracy metrics were extracted and pooled using a bivariate random effects model and hierarchical summary receiver operating characteristic (HSROC) analysis. Subgroup analyses and meta-regression were performed to explore sources of heterogeneity. ResultsEighteen studies published between 2023 and 2025 were included. In title and abstract screening, the pooled sensitivity was 0.92 and pooled specificity was 0.94. The SROC area under the curve (AUC) reached 0.98. In full-text screening, pooled sensitivity and specificity both reached 0.99 and the AUC was 0.99. Prompt strategies incorporating examples or chain-of-thought reasoning significantly improved sensitivity. Across studies, most models were deployed without task specific fine tuning and still achieved strong performance. Subgroup analyses and meta regression did not identify significant sources of heterogeneity. Many studies also reported substantial efficiency gains, including large reductions in screening workload, time and cost. ConclusionLLMs demonstrate high diagnostic accuracy for automated medical literature screening, particularly in full-text assessment. These models show strong potential as high sensitivity assistive tools that can substantially reduce manual screening burden while supporting evidence synthesis. Further methodological optimization and validation in large scale real-world settings are required to establish their long term role in evidence-based medicine.