Artificial intelligence (AI) is rapidly evolving, offering an expanding suite of capabilities that go beyond the traditional focus on prediction and classification. Generative AI (GenAI) and agentic AI could create transformative practices to support real-world evidence (RWE) generation for health research by streamlining studies, accelerating insights and improving decision-making. However, there is no published overview available describing the range of applications in RWE generation. This review aims to describe where and how genAI and agentic AI are applied across the domains of healthcare research tasks for RWE generation. Additionally, to map applications by tasks and methods across the product lifecycle continuum, and to identify emerging gaps and opportunities.

This Living Scoping Review (LSR) will include studies reporting an application and/or evaluation of genAI or agentic AI applied to one or more RWE generation research tasks. Searches will be conducted in Embase, MEDLINE and additional sources (eg, grey literature). Citations will be independently screened by two human senior reviewers for a substantive training dataset and a commercially available screening algorithm (Robot Screener) will complete screening with a human reviewer. The LSR will include reports of studies (primary or reviews) describing and/or evaluating the application of any genAI model for RWE generation in healthcare, in English, published from 1 January 2025 to the date of search. Data will be extracted from all studies included in the LSR by one independent senior reviewer using a piloted template, with 10% quality check by a second senior reviewer. Descriptive statistics will be used to summarise the applications of genAI per RWE research task, and the results of genAI evaluations. Thematic analysis will be used to describe genAI application patterns, trends, gaps and opportunities. The LSR protocol and reports will be updated annually, and findings will be published on a publicly available website (eg, ISPE—the International Society for Pharmacoepidemiology).

Ethical approval is not required due to use of previously published data. Planned dissemination includes peer-reviewed publication, presentation and short summaries.

For people whose stroke risk would be reduced by taking a long-term oral anticoagulant (OAC), it is important to implement effective strategies to support medication initiation, adherence and persistence. To do this, a better understanding of the factors associated with implementation of interventions to optimise OAC management is needed.

This scoping review aimed to summarise the evidence-based characteristics associated with implementing interventions designed to optimise long-term OAC adherence.

Primary research (published post-2000) evaluating any intervention designed to optimise implementation of long-term OAC for stroke prevention by way of change in OAC services, staff or patient behaviour.

Five databases (MEDLINE, Embase, Cumulative Index to Nursing and Allied Health Literature (CINAHL), PsycInfo, Cochrane Library) were searched from 1 January 2000 to 4 August 2023 using a combination of terms relating to population, intervention and study design.

Titles/abstracts were screened by at least one reviewer. Data from each full text were abstracted (with 20% double-checked for accuracy) and its implementation content reviewed, guided by the Expert Recommendations for Implementing Change strategies.

216 studies were included, with varying descriptive reporting of implementation strategies, and only 61 (28%) self-identifying as an implementation study. The median number of implementation strategies used was three, with recently published studies (2015 onwards), those including patients receiving either direct OACs (DOACs) or vitamin K antagonists (VKAs) and those including multiple intervention targets (service, staff or patients) associated with using more implementation strategies. ‘Train and educate stakeholders’ strategies were the most commonly used, and ‘Adapt and tailor to the context’ strategies were the least used by included studies. Conversely, self-defined implementation studies were less likely to use ‘Train and educate stakeholders’ strategies, although they were positively associated with use of ‘Adapt and tailor to the context’. ‘Use evaluative & iterative’ strategies were used more frequently in studies where patients used either VKAs or DOACs, or were published more recently.

Studies need to self-define as implementation studies, improve implementation strategy reporting and be transparently registered, alongside conducting process evaluations or more richly describing implementation processes. Future research could explore why some implementation strategies are used more than others and whether aligning strategy clusters with intervention targets results in clinically significant differences in patient care.

The Veterans Health Administration (VA) integrated mental and physical health services to better detect and treat depression. Primary care nurses conduct screening annually. Clinicians, including Primary Care Mental Health Integration (PCMHI) specialists, follow-up as needed for treatment. Depression detection and management processes are complex, involve multilevel stakeholders, and are subject to significant disruption from COVID-19 and from the resulting expansion of telehealth, aiming to preserve care access. This study aimed to examine whether the COVID-19 pandemic worsened depression-related care quality and/or patient outcomes (eg, suicide).

Given hypothesised care disruption (lowered care quality) during COVID-19, we will first assess the VA population’s trajectory from a new positive depression (and suicide risk) screen to appropriate treatment (ie, medication, therapy) in the Fiscal Year 2019–2323. We will also examine the changing mix of virtual and in-person depression care delivered. Second, we will use interrupted time series analyses to explore the extent to which psychiatric emergency visits and hospitalisations may be mitigated by clinician detection of depression. As well as compare mental health-related mortality rates between patients detected and not detected to have depression. Subanalyses will reveal where (eg, clinics with low PCMHI access) and for whom (eg, minorities) detection does not systematically occur, and downstream negative sequelae, to guide future intervention. Finally, we will interview 40 veterans, half of whom were detected and half not detected to have depression and 40 VA primary care and PCMHI providers about changes brought on by the pandemic and the expansion of virtual care across three VA facilities. In addition to contextualising disrupted care findings, qualitative data will help identify best practices on patient-to-provider and provider-to-provider interactions in hybrid in-person/telehealth depression care models.

Ethics approval was granted by the VA Greater Los Angeles Healthcare System Institutional Review Board. Alongside journal publications, dissemination activities include briefings to our policy and operational partners, and presentations to clinical, research and policy-oriented audiences.

The rapidly growing population of older adults (individuals aged 65 years and older) presents a new set of challenges for healthcare providers in the emergency department (ED), given the prevalence of severe and life-threatening conditions among this group, such as chronic cancer, Alzheimer’s disease/dementia and congestive heart failure. ED encounters often represent a critical point in an older patient’s trajectory of care and can thus be an important opportunity for various interventions such as palliative care consultation. Therefore, identifying those who will benefit most from palliative care is of high importance, especially in determining the course of future treatment. Thus, we aim to conduct a systematic review assessing the efficacy of palliative care screening in the ED by assessing inpatient length of stay as the primary outcome and quality of life, percentage of hospitalisation and cost of care as secondary outcomes.

This study will use Ovid MEDLINE, Embase, EBSCO CINAHL, Web of Science and Cochrane as databases. The study population comprises adults aged 60 years and older, with no focus on any specific clinical specialty or disease. Patients who have not received palliative care screening will serve as the comparator. Only studies with an applicable comparator will be considered. Studies published from 1 January 2000 to 1 July 2025 will be included.

All articles will be reviewed independently and in duplicate, and every author will participate in the review, data abstraction and conflict resolution process.

Ethical approval is not required as it is a protocol for a systematic review. Findings will be disseminated through peer-reviewed publications and conference presentations.

CRD42024562389.

Giant cell arteritis (GCA) is a large-vessel vasculitis occurring in people aged over 50 years. Recent studies have shown that tocilizumab (TCZ), an anti-IL-6 receptor monoclonal antibody, is remarkably effective in treating GCA and allows significant dose sparing of glucocorticoids. However, it makes it difficult to monitor disease activity. Furthermore, treatment is often prolonged over 1 year due to the fear of relapse after stopping TCZ and/or the absence of an optimal discontinuation scheme.

This study aims at comparing two discontinuation regimens in a population of GCA patients who have been treated with TCZ for 12–36 months and have discontinued glucocorticoids for at least 12 weeks. Patients will be randomised with a 1:1 ratio between two arms: immediate discontinuation (cessation) versus gradual discontinuation of TCZ (162 mg subcutaneously every 2 weeks for 12 weeks and then every 4 weeks for 12 additional weeks). Patients will be followed up for 78 weeks. The primary endpoint is relapse-free survival after 26 weeks of follow-up. A total of 120 patients will be randomised (60 in each group) for a period of 3 years.

The trial was approved by an independent ethics committee (CPP Sud Ouest et Outre Mer IV) and the French health authority (French National Agency for Medicines and Health Products Safety—ANSM) through the Clinical Trials Information System (CTIS) provided by the European Medicines Agency (EMA). The informed consent complies with the ICH GCP guideline and regulatory requirements. Eligible patients may only be included in the study after providing informed consent. Findings will be published in peer-reviewed journals and conference presentations.

NCT06037460.

Perineal trauma is one of the most common complications of childbirth, impacting approximately 9 out of 10 women who undergo a vaginal delivery. Perineal trauma is a public health issue leading to increased maternal morbidity and decreased quality of life. Although race is being studied as a potential risk factor and predictor of perineal trauma, other contributing factors like racism and social determinants of health have not been adequately studied in the same context. We set out to synthesise the available peer-reviewed evidence evaluating the prognostic association between race and perineal trauma.

This systematic review and meta-analysis adheres to the PRISMA-P (Preferred Reporting Items for Systematic Reviews and Meta-Analyses Protocols) and PROGRESS (Prognosis Research Strategy) guidelines and is registered with PROSPERO. The review explores the association between racial status (non-Hispanic white vs non-white) and perineal birth trauma using the PECOTS (Population, Intervention/Exposure, Comparator, Outcome, Timing and Setting) framework. We will search PubMed, CINAHL, Web of Science and Embase. Peer-reviewed observational studies will be included. Data extraction and screening will be done in duplicate. Analyses will use random-effects models in R, reporting both unadjusted and adjusted risk differences. Risk of bias will be assessed using ROBINS-I (Risk of Bias in Non-randomised Studies of Interventions). Heterogeneity and certainty of evidence will be evaluated using I² and GRADE (Grading of Recommendations Assessment, Development and Evaluation), respectively.

This is a systematic review based on previously published data, and therefore ethical approval is not required. The findings of this review will be disseminated through publication in a peer-reviewed journal and presented at academic conferences.

CRD42025590093.

Hospitalisation is one of the most stressful life events for older adults, particularly for those who are pre-frail or frail. Multi-component community-based interventions have the potential to address the complex needs of older adults post-acute care admission. While some available interventions have been developed with end-user engagement, fully involving older people who are pre-frail or frail in the design of interventions has been less common. Multi-component community-based interventions that address the needs of older adults and their care partners with potential implementation barriers informed by healthcare providers, community partners and health system decision makers are needed. This protocol paper describes the planned process of co-designing for older patients discharged into the community, a Post-Acute Care Intervention for Frailty using Information and Communication technology.

The development of a complex multi-component frailty intervention which meets older people’s needs involves several concurrent tasks and methodologies, each informed by co-design and conducted with consideration to eventual implementation. These tasks include: (1) establishing a Research Advisory Board, (2) assessing the feasibility and validity of using hospital administrative data to identify frail or pre-frail older adults and their needs, (3) conducting a needs assessment of patients returning to the community, (4) mapping community assets to identify existing programmes and services to help tailor the intervention, (5) co-designing a multicomponent frailty intervention, (6) selecting study outcome measures and (7) selecting and tailoring a digital health patient portal to support intervention delivery, data capture and communication.

Each task requiring ethics approval will be submitted to the Hamilton Integrated Research Ethics Board at McMaster University. Results will be disseminated through peer-reviewed journal articles, conferences and networks of relevant knowledge users who have the capacity to promote dissemination of the results. A toolkit will be developed to help researchers and healthcare providers replicate the methodology for other populations.

To estimate the impact of COVID-19 infection on the requirement for social care services among adults aged ≥50 years in North-West London.

Population-based matched cohort study using linked routinely collected electronic social care, primary care and hospital records (the Discover dataset).

Approximately 4.7 million people with a general practitioner record in North-West London.

150 654 adults aged ≥50 years with a first diagnosis of COVID-19 between January 2020 and February 2023 and 547 704 propensity score matched comparators without a COVID-19 diagnosis during the same period.

Social care use and associated costs overall and by specific type (care home, domiciliary care, respite care, social care assessments) stratified by age group, index year, diagnosis setting, severe COVID-19 risk status, frailty and care home admission prior to index. Overall survival was also assessed.

A total of 9174 (6.09%) individuals with COVID-19 required social care use (of any type) during follow-up, 2.54 times (95% CI 2.48 to 2.61; p

This increase in social care utilisation was observed for all age groups. Adults with COVID-19 had over four times higher social care costs than matched comparators (£1276 per person per year (pppy) vs £276 pppy; mean difference +£1000, 95% CI £947 to £1054, p

COVID-19 infection is associated with meaningfully higher social care requirements in the ≥50 years population. Reducing the need for social care use and the associated costs of care should be one of the goals of interventions to reduce the risk and severity of COVID-19 infection.