Machine learning methods, applications and economic analysis to predict heart failure hospitalisation risk: a scoping review

Background
 Machine Learning (ML) has been transformative in healthcare, enabling more precise diagnostics, personalised treatment regimens and enhanced patient care. In cardiology, ML plays a crucial role in risk prediction and patient stratification, particularly for heart failure (HF), a condition affecting over 64 million people globally and imposing an economic burden of approximately $108 billion annually. ML applications in HF include predictive analytics for risk assessment, identifying patient subgroups with varying prognoses and optimising treatment pathways. By accurately predicting the likelihood of hospitalisation and rehospitalisation, ML tools help tailor interventions, reduce hospital visits, improve patient outcomes and lower healthcare costs.

Objective
To conduct a comprehensive review of existing ML models designed to predict hospitalisation risk in individuals with HF.

Methods
A database search including PubMed, SCOPUS and Web of Science was conducted on 31 March 2024. Studies were selected based on inclusion criteria focusing on ML models predicting hospitalisation risks in adults with HF. The data from 27 studies meeting the criteria were extracted and analysed, with a focus on the predictive performance of the ML models and the presence of economic analysis.

Results
Most studies focused on predicting readmission rather than first-time hospitalisation. All included studies employed supervised ML algorithms, with ensemble-based methods generally yielding the highest predictive performance. For 30-day hospitalisation or readmission risk, Extreme Gradient Boosting (XGBoost) achieved the highest mean area under the curve (AUC) (0.69), followed by Naïve Bayes (0.68) and Deep Unified Networks (0.66). For 90-day risk, the best-performing models were Least Absolute Shrinkage and Selection Operator and Gradient Boosting, both with a mean AUC of 0.75, followed by Random Forest (0.67). When the prediction timeframe was unspecified, Categorical Boosting achieved the highest performance with a mean AUC of 0.88, followed by Generalised Linear Model Net and XGBoost (both 0.79).
Electronic health records were the primary data source across studies; however, few models included patient-reported outcomes or socioeconomic variables.
None of the studies conducted an economic evaluation to assess the cost-effectiveness of these models.

Conclusions
ML holds substantial potential for improving HF care. However, further efforts are needed to enhance the generalisation of models, integrate diverse data sources and evaluate the cost-effectiveness of these technologies.

Read More

Exploring evidence gaps in clinical trials in thermal burns care: an umbrella review

Background
Gaps in research evidence lead to research waste. In burns treatment, there is a paucity of reliable evidence or data. This contributes to inconsistent patient care, especially on a global scale, where low-resource countries often lack access to the latest research advancements. This umbrella review was undertaken as part of the James Lind Alliance Priorities in Global Burns Research Prioritisation Setting Partnership (PSP) and aimed to identify and assess the quality of evidence in thermal burns care. The objective was to map which interventions in thermal burns care are supported by a reliable evidence base and for which the evidence is lacking.

Methods
Systematic reviews of randomised controlled trials in thermal burns were identified and assessed using reliability criteria determined a priori. Multiple systematic review databases were searched in June 2023, including the Cochrane Library, KSR Evidence database and NIHR Journals Library. Summary of findings and, where available, Grading of Recommendations Assessment, Development and Evaluation was used to assess certainty of evidence. Reliable reviews were mapped onto clinical categories identified by patients, carers and healthcare professionals as part of the PSP.

Results
232 systematic reviews were identified, of which 83 met reliability criteria and were included. The main reason for not meeting reliability criteria was poorly defined eligibility criteria (n=128). Of the 83 reliable reviews, most were conducted in pain (n=28) or wound management (n=14) and acute care (n=13). Certainty of evidence was mixed. Reviews mapped onto nine of the 17 clinical categories identified by the PSP.

Conclusion
This review summarises the available high-quality evidence in burns care and identifies evidence gaps, indicating that many important clinical questions remain unanswered. There is a discrepancy between the treatments investigated in high-quality research and the clinical areas considered as most important to stakeholders. These findings provide direction for future research to improve global burns care.

Read More

Development and validation of critical appraisal tool for individual participant data meta-analysis: protocol for a modified e-Delphi study

Introduction
Individual participant data meta-analysis (IPD-MA) is regarded as the gold standard for evidence synthesis. However, diverse recommendations and guidance on its conduct exist, and there is no consensus-based tool for the critical appraisal of a completed IPD-MA. We aim to close this gap by systematically identifying quality items and developing and validating a critical appraisal checklist for IPD-MA.

Methods and analysis
This study will comprise three phases, as follows:
Phase 1: a systematic methodology review to identify potential checklist domains and items; this will be conducted according to the Cochrane methods for systematic reviews and reported following the Preferred Reporting Items for Systematic Reviews and Meta-analysis 2020 guidance. We will include studies that address methodological guides and essential statistical requirements for IPD-MA. We will use the proposed items to prepare a preliminary checklist for the e-Delphi study.
Phase 2: at least two rounds of an e-Delphi survey will be conducted among panels with expertise in IPD-MA research, consensus development, healthcare providers, journal editors, healthcare policymakers, patients and public partners from diverse geographic locations with experience in IPD-MA. Participants will use Qualtrics software to rate items on a 5-point Likert scale. The Wilcoxon matched signed rank test will estimate response stability across rounds. Consensus on including an item will be achieved if ≥75% of the panel rates the item as ‘strongly agree’ or ‘agree’ and items will be excluded if ≥75% rates it as ‘strongly disagree’ or ‘disagree’. A convenience sample of 10 reviewers with experience in conducting an IPD-MA will pilot-test the checklist to provide practical feedback that will be used to refine the checklist.
Phase 3: critical appraisal checklist validation: to improve confidence in the tool’s uptake, a subset of the e-Delphi participants and graduate students of epidemiology and biostatistics will conduct content validity and reliability testing, respectively, per the Consensus-based Standards for the Selection of Health Measurement Instruments.

Ethics and dissemination
Ethics approval has been obtained from the Western University Health Science Research Ethics Board in Canada. The validated checklist will be published in a peer-reviewed open-access journal and shared across the networks of this study’s steering committee, Cochrane IPD-MA group and the institutions’ social media platforms.

Read More

Effectiveness of education and training programmes to help clinicians assess and classify perineal tears: a systematic review

Objective
For every 100 women having a vaginal birth, approximately four will experience a severe (third-degree or fourth-degree) perineal tear. Severe perineal tears are associated with significant short-term and long-term consequences if not recognised and repaired. There are global efforts to reduce incidence of severe perineal tears including initiatives to strengthen education and training of clinicians in perineal anatomy and perineal tear assessment and classification. This systematic review aimed to describe and evaluate the effectiveness of these initiatives.

Design
Systematic review, reported in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines.

Data sources
Medline (Ovid), CINAHL Complete (EBSCO), MIDIRS and EMBASE (Ovid) were searched through 15 February 2024.

Eligibility criteria
Studies reporting simple or complex interventions aimed at improving the skills and knowledge of clinicians in perineal anatomy and/or the clinical assessment and classification of perineal tears were eligible.

Data extraction and synthesis
Two reviewers independently screened studies for eligibility and appraised the quality of individual studies using the Cochrane Risk of Bias (RoB) 2.0 tool or Risk of Bias in Non-Randomised Studies of Interventions (ROBINS-I) tool.

Results
In total, 7645 citations were screened and 39 studies included. We identified nine unique interventions including training programmes, short workshops and senior obstetrician supervision. Many studies were from high-income countries, in primary care settings and at high risk of bias.

Conclusions
Effective education included practical components, such as hands-on skills and training in perineal anatomy, assessment and classification, rather than senior supervision alone. Ongoing review of practice appears to be crucial for maintaining knowledge and skills. Future research should focus on interventions tailored to limited-resource settings, and the optimal length and intensity of training programmes to assess and classify perineal tears.

PROSPERO registration number
CRD42020185431.

Read More

Comparing non-ischaemic heart preservation (NIHP) with ischaemic static cold storage of donor hearts in adult cardiac transplantation: study protocol for a randomised controlled trial

Introduction
Ischaemia–reperfusion (I/R) injury remains a major challenge in heart transplantation, with mortality risk increasing significantly when allograft ischaemic time exceeds 4 hours. Non-ischaemic heart preservation (NIHP), using continuous hypothermic perfusion, has shown promise in preliminary studies for reducing I/R injury and improving outcomes. This randomised controlled trial aims to compare NIHP with standard static cold storage (SCS) in adult heart transplantation.

Methods and analysis
The trial is a prospective, open-label, multicentre, single-blinded, randomised controlled trial including 66 adult heart transplant recipients across four Swedish hospitals. Participants will be randomised into 1:1 ratio to NIHP or SCS preservation groups and undergo a 12-month follow-up period. The primary outcome is 1-year survival free from acute cellular rejection or retransplantation. Secondary outcomes include quality of life, I/R injury markers, graft function and adverse events. Substudies will evaluate renal function using MRI and continuously monitor physical activity and heart rhythm via wearable devices. Analysis will follow intention-to-treat principles, with time-to-event analysis using Cox proportional hazard models and Kaplan-Meier estimates.

Ethics and dissemination
The study has been approved by the Swedish Ethical Review Authority. It will be conducted according to the Declaration of Helsinki and relevant local and international regulations. Results will be published in peer-reviewed journals following Consolidated Standards of Reporting Trials guidelines.

Trial registration number
NCT04066127.

Read More

Brain-Body Interactions in Ischemic Stroke: VNS Reprograms Microglia and FNS Enhances Cerebellar Neuroprotection

Stroke, Ahead of Print. Stroke significantly impacts mortality and long-term disability, necessitating effective rehabilitation strategies to enhance recovery. This review examines the roles of vagus nerve stimulation (VNS) and fastigial nucleus stimulation (FNS) in facilitating ischemic stroke recovery through brain-body interactions. VNS enhances ischemic stroke recovery by reprogramming microglia from proinflammatory (M1) to neuroprotective (M2) phenotypes, reducing neuroinflammation and promoting tissue repair via neurotrophic factors. It has shown promise in clinically improving chronic upper limb deficits when combined with rehabilitation therapies. Conversely, FNS provides cerebellar-mediated neuroprotection by mainly mitigating excitotoxic damage and inflammatory responses independent of cerebral blood flow alterations, as evidenced by preclinical models of middle cerebral artery occlusion. By integrating VNS-driven immunomodulation with FNS-mediated excitotoxicity suppression, this review highlights their synergistic potential to improve rehabilitation outcomes for ischemic stroke survivors. Biomarker-guided protocols: VNS for cortical/subcortical ischemic deficits and FNS for cerebellar network recovery are advocated to address postischemic disability via anti-inflammatory rewiring, neuroplasticity enhancement, and cerebellar-thalamocortical circuit stabilization. Critical gaps remain in hemorrhagic stroke, where FNS’s excitotoxicity suppression may destabilize clots, necessitating subtype-specific safety validations.

Read More

Evaluation of the outpatient parenteral antimicrobial therapy (OPAT) service in Malaysian public hospitals: a mixed-methods study protocol

Introduction
Outpatient parenteral antimicrobial therapy (OPAT) is an innovative approach to manage infections that require extended courses of intravenous antibiotics by enabling patients to receive treatment in an outpatient setting. In Malaysia, there has yet to be a systematic evaluation of the OPAT service. This study aims to describe the safety, clinical indications and treatment outcomes of the OPAT service in Malaysia, assess patients’ satisfaction and experiences and determine the facilitators and barriers associated with the provision of the OPAT service in Malaysia.

Methods and analysis
A mixed-methods approach combining qualitative and quantitative methods will be employed for a comprehensive understanding of the provision of the OPAT service in Malaysian public hospitals. The study consists of four distinct parts: systematic review, retrospective cohort analysis of clinical outcomes, patients’ satisfaction survey and focus group discussions on providers’ experiences. A longitudinal analysis of the clinical outcomes (treatment success/failure, infection cure, adverse events, readmission and mortality) of the OPAT patients’ cohort will be conducted using descriptive and conclusive statistics, in addition to rates of patients’ satisfaction and evaluation of providers’ experiences.

Ethics and dissemination
This study is registered in the National Medical Research Register (NMRR ID-24-00941-2C8) and approved by the Medical Research and Ethics Committee, Ministry of Health Malaysia (Ref: 24-00941-2C8). Written informed consent will be obtained from all participants. The results will be submitted for publication in a peer-reviewed journal and presented at one or more scientific conferences.

Read More

Randomised trial of home sleep apnoea testing compared to in-lab polysomnography for the evaluation of obstructive sleep apnoea in children: rationale and study protocol

Introduction
Obstructive sleep apnoea (OSA) affects 1–5% of the paediatric population, including 55–90% of children with Down syndrome (DS), and has been associated with negative effects on neurocognitive development, cardiovascular health, immune development and quality of life. In-lab attended polysomnography (PSG) is currently the gold standard for the diagnosis of OSA in children, but it poses challenges due to the burden on families and limited testing facilities. Home sleep apnoea testing (HSAT), an unattended sleep test done at home, is an accepted alternative for adults but lacks sufficient evidence to be used clinically for the evaluation of OSA in children. HSAT may be especially beneficial for children with DS or others with sensory issues or those who struggle with sleeping in a laboratory setting overnight.

Methods and analysis
This single-centre trial compares HSAT to PSG for the diagnosis of OSA in children, including those with DS. The trial will enrol 317 children 5–12 years old, including approximately 100 with DS. The primary outcome is the diagnostic accuracy of HSAT compared with PSG for OSA evaluated through ROC. Secondary outcomes include the agreement between HSAT and PSG for therapeutic decision-making and comparison of preference and acceptability of HSAT versus PSG. This trial seeks to evaluate HSAT as an alternative diagnostic tool for paediatric OSA, potentially expanding testing options for clinicians and families.

Ethics and dissemination
This study has been approved by the Institutional Review Board at Children’s Hospital of Philadelphia (#21–0 19 533). Informed consent will be obtained from all participants, and no identifiable data will be reported.

Trial registration number
NCT05382754.

Read More

Effect of transcranial magnetic stimulation on perioperative neurocognitive dysfunction: study protocol for a systematic review and meta-analysis of randomised controlled trials

Introduction
Perioperative neurocognitive disorders (PNDs), a common postoperative complication associated with anaesthesia and surgical procedures, are characterised by impairments in memory, attention, language comprehension and social functioning. Accumulating evidence from clinical studies indicates that transcranial magnetic stimulation (TMS)—a non-invasive neuromodulatory modality capable of targeted cortical stimulation—may offer therapeutic promise for PND management. To comprehensively assess the intervention efficacy and safety parameters of TMS in mitigating postoperative cognitive decline, we propose conducting a systematic review and meta-analysis of randomised controlled trials adhering to Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines. Our findings aim to provide evidence-based insights into the neuroprotective potential of TMS for mitigating cognitive decline in surgical populations.

Methods and analysis
The investigation will implement a multifaceted search protocol encompassing international and Chinese scholarly resources. The search will be conducted in major databases, including Web of Science, Embase, PubMed, Cochrane Library, China Biology Medicine, China National Knowledge Infrastructure, Wan Fang Database and Chinese Scientific Journal Database from inception to 31 March 2025. To capture the latest research trends, ongoing trials will be simultaneously searched in the WHO International Clinical Trial Registry Platform, ClinicalTrials.gov and the China Clinical Trial Registry. Grey literature will be supplemented through resources such as GreyNet International, OpenGrey and Google Scholar. The inclusion criteria for this review are restricted to randomised controlled trials investigating the application of TMS as an intervention for PND. Primary endpoints comprise clinically confirmed incidence rates of postoperative delirium and delayed neurocognitive recovery. Two researchers will independently perform literature screening, data extraction and risk of bias assessment. The risk of bias in included studies will be evaluated using the Cochrane Risk of Bias Tool 2.0. Evidence certainty will be appraised through the GRADE framework with explicit justification for downgrading decisions. Meta-analysis will be conducted using STATA V.15.1 statistical software. The data synthesis process will incorporate standardised methodologies, including heterogeneity testing, sensitivity analysis and assessment of publication bias.

Ethics and dissemination
This study will not involve the collection of biometric information or medical privacy data throughout the research process, thus complying with the exemption criteria outlined in the ‘Measures for Ethical Review of Biomedical Research Involving Human Subjects.’ The findings will adhere to academic standards and be submitted for publication in reputable international medical journals following a rigorous double-blind peer-review process.

Systematic review registration
The research protocol has been prospectively registered on the PROSPERO international prospective systematic review registration platform (registration number: CRD42025636978).

Read More

The emerging economic evidence and methods used to evaluate clinical registries: a systematic scoping review protocol

Introduction
A clinical registry is a systematically collected database of health-specific information about a patient population. Clinical registries can be used for a variety of purposes including surveillance, monitoring of outcomes and patient care. The establishment and maintenance of clinical registries come with a significant cost. This scoping review aims to identify the methods used to economically evaluate clinical registries including their costs and benefits.

Methods
This systematic scoping review protocol has been developed in accordance with the Preferred Reporting Items for Systematic Review and Meta-Analysis Protocols (PRISMA-P) guidelines. The final review will be reported following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses extension for Scoping Reviews (PRISMA-ScR) checklist. The electronic databases Medline, Embase, Cochrane Library and The Cumulative Index to Allied Health Literature(CINAHL) database will be searched. Relevant national organisation websites will be searched to identify empirical studies within grey literature. The inclusion criteria include studies that economically evaluate clinical registries and are published in the English language from inception to February 2025. Two reviewers will independently screen 100% of titles and abstracts and full texts of studies for inclusion. Data will be extracted from eligible studies prior to being assessed for quality using a multi-tool approach.

Ethics and dissemination
The findings of this review will be published in an international peer-reviewed journal. They are likely to be of interest to custodians of existing clinical registries and to those wishing to establish or evaluate clinical registries.
Keywords
Clinical registries, economic evaluation, costs, cost-effectiveness, health economics, registry based studies

Read More

Self-administered dual-task training reduces balance deficits and falls among community-dwelling older adults: a multicentre parallel-group randomised controlled trial with economic evaluation protocol

Background
Falls are common causes of disability, reduced disability-adjusted life years and death in older adults. Balance deficits and cognitive impairment are common causes of falls. Dual-task training is a new strategy that can potentially improve balance and cognitive function, leading to decreased falls. The effectiveness and cost-effectiveness of self-administered dual-task (sDT) training to improve balance and prevent falls is not known. We developed sDT training combining physical and cognitive tasks to improve balance and reduce falls. The proposed randomised controlled trial (RCT) with economic evaluation is to test the effectiveness and cost-effectiveness of the sDT compared with self-administered single-task training (sST) in this population.

Methods and analysis
In this RCT, we will recruit 190 community-dwelling older adults with a history of at least one fall over the last 6 months from 11 elderly centres. The older adults will be randomly assigned to the sDT (n=95) and sST groups (n=95). Each group will be offered in six training workshops to teach the participants either sDT or sST depending on the group allocation. Each workshop will last an hour and will be held once every 2 weeks for 3 months. Besides, the participants will be instructed to repeat the exercises at home two times weekly for 3 months. Following the intervention phase, the participants will continue unsupervised home-based exercises for 6 months. Assessments will be performed before, after and 6 months after completing the intervention. A fall calendar and cost diary will be provided to each older adult to record the number of falls and fall-related costs during and after the intervention to assess fall incidence and cost-effectiveness. Effectiveness will be assessed using a negative binomial regression model following the intention-to-treat principle for falls and a linear mixed model for the additional measure and cost-effectiveness using a Markov model.

Ethics and dissemination
This study has ethical approval from the PolyU Institutional Review Board for conducting research on human subjects (Ref: HSEARS20210322005). The results will be disseminated through seminars for individuals and health practitioners, international conferences and published in peer-reviewed journals.

Trial registration number
NCT05533333.

Read More

Burden of disease using disability adjusted life years in the Middle East and North Africa (MENA) region: protocol of a systematic review

Introduction
In the Middle East and North Africa (MENA) region, changing demographic and epidemiological profiles have resulted in a diverse and shifting burden of disease (BoD). Disability-adjusted life years (DALYs), which combine years of life lost (YLL) due to premature mortality and years lived with disability (YLD), offer a valuable metric for assessing disease burden at the national level. While global burden of disease (GBD) estimates provide broad insights, national burden of disease (NBD) estimates offer country-specific data that can better inform tailored health policies and resource allocation. This systematic review protocol outlines our methodology for collating and analysing the NBD estimates in the MENA region using DALYs as the primary outcome measure.

Methods and analysis
This review will adhere to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. We will systematically search PubMed, Scopus, Web of Science and EMBASE for studies published from 1993 to 2024 that report national-level DALY estimates for diseases, injuries or risk factors in MENA countries. Eligible studies must report DALY estimates using national methodologies, while studies using exclusively GBD estimates will be excluded. Two independent reviewers will conduct title/abstract and full-text screening, data extraction and quality assessment using Standardised Reporting of Burden of Disease Studies (STROBOD), with disagreements resolved by a third reviewer when necessary.

Ethics and dissemination
Ethical approval is not required for this review as it involves analysis of previously published data. The findings will be disseminated through publication in a peer-reviewed journal and presented at relevant academic and policy forums.

Prospero Registration Number
PROSPERO CRD42024498688.

Read More

Development of core competency evaluation index system for informatics nurses in China: a Delphi study

Objective
To develop a scientific and systematic core competency evaluation index system for informatics nurses in China. The goal is to support their training, assessment and performance evaluation.

Methods
An initial set of evaluation indicators was created based on a review of existing literature and semi-structured interviews. From September to December 2023, the indicators were refined through two rounds of Delphi expert consultation. A total of 21 experts from 11 provinces in China completed the questionnaires.

Results
The response rate for both rounds of questionnaire surveys was 100.0%. The expert authority coefficient was 0.934. The coordination coefficient among experts was statistically significant (p

Read More