Septic shock is a severe complication of acute pancreatitis (AP), often associated with poor prognosis. This study aims to analyze the clinical characteristics of patients with acute pancreatitis and develop an interpretable early prediction model for septic shock in these patients using machine learning (ML). The model is intended to assist emergency physicians in resource allocation and medical decision making. Data were collected from the MIMIC-IV 3.0 database. The dataset was divided into a training set and a test set in a 7:3 ratio. Feature selection was performed using LASSO (Least Absolute Shrinkage and Selection Operator) regression. Subsequently, 10 ML models were developed: Random Forest, Logistic Regression, Gradient Boosting Machine, Neural Network, Extreme Gradient Boosting (XGBoost), K-Nearest Neighbor, Adaptive Boosting, Light Gradient Boosting Machine, Category Boosting, and Support Vector Machine. To enhance and optimize model interpretability, Shapley Additive Explanations (SHAP) were employed. A total of 1032 patients with AP were included in this study, from which 31 variables were selected for model development. By comparing the area under the receiver operating characteristic curve and decision curve analysis results between the training and test sets, the XGBoost model demonstrated a significant advantage over other models. SHAP analysis revealed that white blood cell count, total bilirubin (bilirubin total), and bicarbonate (HCO3 -) levels were the three most critical risk factors for the development of septic shock in patients with AP. ML approaches exhibited promising performance in predicting septic shock in patients with AP. These models may aid in guiding treatment decisions for patients with AP in the emergency department.

BackgroundThe Internet has been integrated into various aspects of senior life. However, the popularity of digitization does not mean that digital devices are well-inclusive. It is considered important to explore the willingness of nursing home residents to use smartphones in the context of the development of “Digital China.”ObjectivesTo explore the perceptions and concerns of smartphone use in nursing home residents.MethodThe study employed a descriptive qualitative design. A purposive sample of 22 nursing home residents was recruited as participants. Semi-structured interviews were conducted to collect data between April 2024 and June 2024. Interviews were transcribed, and analyzed by the qualitative content analysis method.ResultsNursing home residents were wandering at the crossroads of smartphone use. Four specific categories emerged: (1) Positive appraisal of smartphones’ value; (2) Internal cognitive biases in smartphone use; (3) Objective constraints on smartphone use; and (4) Practical strategies for digital inclusion. Their cognitive biases involved low perceived usefulness, lack of basic knowledge, and technological anxiety; and the objective constraints were physical aging, unfriendly operating systems, and poor digital inclusion environments. Residents expected instrumental digital support including visual smartphone use guidelines, peer or intergenerational beside guidance, and digital life scenarios.ConclusionsThis study clarifies the subjective and objective challenges encountered by nursing home residents in using smartphones. It provides fresh insights from the perspective of nursing home residents for developing environmentally sensitive digital inclusion intervention strategies.

BackgroundAs a representative product of generative artificial intelligence (GenAI), ChatGPT demonstrates significant potential to enhance healthcare outcomes and improve the quality of life for healthcare consumers. However, current research has not yet quantitatively analysed trust-related issues from both the healthcare consumer perspective and the uncertainty perspective of human–computer interaction.ObjectiveThis study aims to analyse the antecedents of healthcare consumers’ trust in ChatGPT and their adoption intentions towards ChatGPT-generated health information from the perspective of uncertainty reduction.MethodsAn anonymous online survey was conducted with healthcare customers in China between September and October 2024. This survey included questions on critical constructs such as social influence, situational normality, anthropomorphism, autonomy, personalisation, information quality, information disclosure, trust in ChatGPT, and intention to adopt health information. A 7-point Likert scale was used to score each item, ranging from 1 (strongly disagree) to 7 (strongly agree). SmartPLS 4.0 was used to analyse data and test the proposed theoretical model.ResultsThe findings indicated that trust in ChatGPT had a significant relationship with the intention to adopt health information. The primary factors associated with trust in ChatGPT and the intention to adopt health information were social influence, situational normality, autonomy, personalisation, and information quality. The analysis revealed a negative relationship between social influence and trust in ChatGPT. Familiarity with ChatGPT was identified as a significant control variable.ConclusionTrust in ChatGPT is positively related to healthcare consumers’ adoption of health information, with information quality as a key predictor. The findings offer empirical support and practical guidance for enhancing trust and encouraging the use of GenAI-generated health information.

BackgroundIn rural states, paediatric capacities and competencies are concentrated in urban centres due to demographic change and economical pressure. This results in greater distances, longer waiting times and multiple examinations for rural-dwelling children seeking healthcare. A regional tele-paediatric network (RTP-Net) was implemented to improve the access of children to paediatric care by providing videoconferencing and digital health information exchange between paediatricians from hospitals with different sizes and degrees of specialization.ObjectiveEvaluation of the user acceptance as well as implementation barriers and facilitators.MethodsBetween January 2021 and March 2024, mixed-methods were carried out including participant observation and a survey of the RTP-Net users. Descriptive analysis based on the Technology Acceptance Model according to Davis. The study report follows ‘A comprehenSive tool to Support rEporting and critical appraiSal of qualitative, quantitative and mixed methods implementation reSearch outcomes’ (ASSESS).ResultsA total of 507 tele-consultations were conducted. 907 questionnaires were analysed. The respondents were satisfied with 92.1% of the tele-consultations. The respondents found most of the tele-consultations (95.8%) adequate for solving their questions. Technical issues were the most mentioned reason (59.3%) why tele-consultation were inadequate; 184 project diary entries were analysed; 42 times (61%), the participants highlighted positive usability aspects of the RTP-Net (e.g., improved availability of patient health data) and 27 times (39%) negative aspects (e.g., interoperability issues).ConclusionsA regional tele-paediatric healthcare network is feasible and acceptable in order to improve paediatric care in rural regions. However, technical improvements and policy adjustments are needed to enhance telemedicine adoption.Trial registration: This study was registered under the title ‘Implementation und Evaluation of a Regional Tele-paediatric Network in Mecklenburg-Vorpommern and Brandenburg' (registration no. DRKS00024002, https://drks.de/search/de/trial/DRKS00024002) at the German Registry for Clinical Trials.

Short video-based social media platforms, such as TikTok, Instagram Reels, and YouTube Shorts, are changing the way individuals’ access and search for health information. This study analyzed the factors that influence the intention and behavior of users in searching for health information on short video platforms. A mixed-methods approach was used by distributing questionnaires to 566 respondents and interviewing 30 respondents. The questionnaire data were analyzed using partial least squares structural equation modeling, and the interview data were analyzed using content analysis. Visual appeal was found to impact the perceived usefulness and the perceived ease of use, while user interactivity only impacted the perceived ease of use. Health consciousness did not influence the intention to search for health information online. The intention to search for health information online was found to significantly influence the actual online health information-seeking behavior. These findings indicate that user intent affects the actual behavior of users in seeking health information. This study contributes to the theoretical understanding of the behavior of searching for health information online and offers insights for platform developers to improve the visual appeal, ease of use, and credibility of their platforms.

ObjectiveThe correlation between health data quality and disease management is of utmost importance. Optimal data quality enhances decision making, minimizes medical errors, and boosts efficiency in disease management. Hence, current research is focused on mapping of co-occurrences and the relationship between health data quality and disease management.MethodsThe present study employed a scientometric approach known as co-occurrence analysis to analyze the data extracted from the Web of Science. Our research community aimed to investigate the association between health data quality and disease management, encompassing all documents produced in this field from 1991 to September 2023. The data was meticulously scrutinized and recorded on 1 October 2023. To generate graphs and tables, we utilized Excel 2019 software utilized, while VOSviewer scientometric software was employed to create co-occurrence maps.ResultsThe top three countries with the highest scientific production in this field were the United States, England, and Australia. The study also uncovered three clusters centered on data quality, management, and public health. Co-occurrence maps drawn from the study showed that accuracy, coverage, and completeness of health data are related to the provision of healthcare and therapeutic interventions, as well as treatment outcomes. Furthermore, management was found to be associated with prevention, epidemic, prevalence, disease, and mortality rate. The study also revealed that data quality is linked to evaluation, validity, reliability, public health, and quality of life.ConclusionIt is clear from this research that accuracy, coverage, and completeness are crucial characteristics of data quality. These factors play a key role in the management of prevention, epidemics, and disease outbreaks, as well as and mortality rates. Evaluating data quality is also important as it can have a positive impact on the quality of life, public health, and health information technologies. This study has the potential to be very useful in improving the quality of health data, the clinical informationist performances, and in turn, promoting more effective disease management.

Diabetes management requires continuous monitoring of glucose, medication, diet, and activity. With over 537 million cases globally, digital health solutions support self-care. This study assesses the Madhumeh Setu app, comparing its functionality, quality, and user experience with top diabetes apps using Mobile App Rating Scale (MARS). A four-phase evaluation was conducted: systematic app search from Google Play Store; assessment of essential app features based on AADE-7 guidelines; qualitative ratings by healthcare professionals using MARS, and user comparison of Madhumeh Setu with MySugr and BeatO through MARS. Out of 2116 screened apps, 11 met the inclusion criteria. MySugr received the highest MARS rating (4.40 ± 0.14), followed by BeatO (4.12 ± 0.04) and Madhumeh Setu (4.05 ± 0.05). While MySugr scored highest overall, Madhumeh Setu performed similarly to BeatO in functionality and engagement. A one-way ANOVA showed that Madhumeh Setu differed significantly from MySugr (p < 0.00001) but was like BeatO, indicating it matches existing diabetes management solutions. This study underscores the potential of digital health solutions in diabetes care. As mobile health advances, culturally specific apps such as Madhumeh Setu will become vital for diabetes management. Future studies should use long-term trials to evaluate their effects on blood sugar, medication use, and well-being, with ongoing improvements based on user input to aid wider adoption, especially in low-and-middle-income countries.

ObjectiveGenerative AI is increasingly used to provide health-related information in addition to online health information seeking (OHIS). Users’ willingness to adopt it is crucial. This study investigates individual factors associated with more frequent OHIS: health status, health anxiety, and eHealth literacy. Using the Technology Acceptance Model (TAM), we examined whether these factors are related to more trust in generative AI for health-related purposes and the willingness to use it.MethodsUsing SEM, we analyzed cross-sectional survey data (N = 4775) that is representative of adult Czech internet users (50% female; aged 18–95 years).ResultsTrust in AI was strongly associated with willingness to use AI. Health status and health anxiety were related to willingness to use AI only indirectly through trust. Higher eHealth literacy was associated with more trust only marginally and had no direct relationship with willingness to use AI. Wellness-related OHIS was positively associated with willingness to use AI for wellness purposes, and illness-related OHIS was associated with willingness to use AI for illness purposes.ConclusionAlthough not emphasized in TAM and its health-related extensions, trust seems to be a critical mediator in the adoption of generative AI for health purposes. Other factors related to OHIS were not associated with willingness to use AI, except for their relationship with trust. eHealth literacy is practically unrelated to trust and willingness to use AI, which is noteworthy given that health anxiety and health status related to higher acceptance are associated with more risky or high-stake use of online health information.

Telemedicine platforms played a crucial role during the COVID-19 pandemic, alleviating issues related to the shortage and unequal distribution of healthcare resources. The purpose of this study is to identify key factors affecting the service quality of telemedicine platforms in China, with the dual objectives of advancing patient wellbeing and informing evidence-based service innovations for industry stakeholders. To quantitatively assess the impact of these key factors on health and wellbeing from the perspective of healthcare consumers, a total of 25,499 valid online reviews were collected from telemedicine platforms. To establish a service quality evaluation framework, this study proposes a novel approach that combines the Servqual quality assessment model with a CNN-BiLSTM deep learning model enhanced by an attention mechanism. Analysis of the full sample shows that healthcare consumers are most concerned about the quality of services provided by telemedicine platforms, with the most important being the professional competence of doctors, a critical factor for promoting consumer health and wellbeing. The proposed hybrid deep learning approach demonstrates superior performance in sentiment classification accuracy, outperforming conventional methods by 11.11 percentage points. This methodological innovation enables more precise identification of consumer sentiment patterns across service dimensions. The novel quality assessment framework introduced here provides actionable insights for advancing telemedicine platforms, driving progress toward precision healthcare and consumer-centric wellbeing. Furthermore, it enables healthcare consumers to select telemedicine services aligned with their personalized needs.

This historical opinion article draws parallels between mid-20th-century psychiatric deinstitutionalization and today's rapid expansion of home-based medical care. While psychiatric conditions differ fundamentally from acute medical illnesses, lessons from deinstitutionalization can inform current and future telemedicine models. Deinstitutionalization resulted from social, legal, and pharmaceutical advances but also produced unforeseen consequences, including increased incarceration, homelessness, medication misuse, and significant family burden. Similarly, shifting acute care to the home may expose patients and caregivers to new stresses: elevated risk of medical complications, uneven access to technology, cost-shifting, and the potential neglect of those most vulnerable or socially disadvantaged. Historical patterns show that successful community-based care demands strong infrastructure, equitable resource allocation, and thorough patient selection. This manuscript urges stakeholders to heed these lessons and build robust, multidisciplinary, and family-centered systems to support patients transitioning to home care. With proper planning, the current dehospitalization process can achieve its promise of high-quality, cost-effective care without repeating the pitfalls of past reforms.