In rural areas, work-integrated learning in the form of health student placements has several potential benefits, including contributing to student learning, enhancing rural health service capacity and attracting future rural health workforce. Understanding what constitutes a high-quality rural placement experience is important for enhancing these outcomes. There is no current standardised definition of quality in the context of rural health placements, nor is there understanding of how this can be achieved across different rural contexts. This study is guided by one broad research question: what do university staff believe are the determinants of high-quality health professions student placements in regional, rural and remote Australia?

This study will adopt a convergent mixed-method design with two components. Component A will use explanatory sequential mixed methods. The first phase of component A will use a survey to explore determinants that contribute to the development of high-quality health student placements from the perspective of university staff who are not employed in University Departments of Rural Health and are involved in the delivery of health student education. The second phase will use semistructured interviews with the same stakeholder group (non-University Department of Rural Health university staff) to identify the determinants of high-quality health student placements. Component B will use a case study Employing COnceptUal schema for policy and Translation Engagement in Research mind mapping method to capture determinants that contribute to the development of high-quality health student placements from the perspective of University Department of Rural Health university staff.

The University of Melbourne Human Ethics Committee approved the study (2022-23201-33373-5). Following this, seven other Australian university human research ethics committees provided external approval to conduct the study. The results of the study will be presented in several peer-review publications and summary reports to key stakeholder groups.

Early identification of lung cancer on chest radiographs improves patient outcomes. Artificial intelligence (AI) tools may increase diagnostic accuracy and streamline this pathway. This study evaluated the performance of commercially available AI-based software trained to identify cancerous lung nodules on chest radiographs.

This retrospective study included primary care chest radiographs acquired in a UK centre. The software evaluated each radiograph independently and outputs were compared with two reference standards: (1) the radiologist report and (2) the diagnosis of cancer by multidisciplinary team decision. Failure analysis was performed by interrogating the software marker locations on radiographs.

5722 consecutive chest radiographs were included from 5592 patients (median age 59 years, 53.8% women, 1.6% prevalence of cancer).

Compared with radiologist reports for nodule detection, the software demonstrated sensitivity 54.5% (95% CI 44.2% to 64.4%), specificity 83.2% (82.2% to 84.1%), positive predictive value (PPV) 5.5% (4.6% to 6.6%) and negative predictive value (NPV) 99.0% (98.8% to 99.2%). Compared with cancer diagnosis, the software demonstrated sensitivity 60.9% (50.1% to 70.9%), specificity 83.3% (82.3% to 84.2%), PPV 5.6% (4.8% to 6.6%) and NPV 99.2% (99.0% to 99.4%). Normal or variant anatomy was misidentified as an abnormality in 69.9% of the 943 false positive cases.

The software demonstrated considerable underperformance in this real-world patient cohort. Failure analysis suggested a lack of generalisability in the training and testing datasets as a potential factor. The low PPV carries the risk of over-investigation and limits the translation of the software to clinical practice. Our findings highlight the importance of training and testing software in representative datasets, with broader implications for the implementation of AI tools in imaging.

One-third of children in England have special educational needs (SEN) provision recorded during their school career. The proportion of children with SEN provision varies between schools and demographic groups, which may reflect variation in need, inequitable provision and/or systemic factors. There is scant evidence on whether SEN provision improves health and education outcomes.

The Health Outcomes of young People in Education (HOPE) research programme uses administrative data from the Education and Child Health Insights from Linked Data—ECHILD—which contains data from all state schools, and contacts with National Health Service hospitals in England, to explore variation in SEN provision and its impact on health and education outcomes. This umbrella protocol sets out analyses across four work packages (WP). WP1 defined a range of ‘health phenotypes’, that is health conditions expected to need SEN provision in primary school. Next, we describe health and education outcomes (WP1) and individual, school-level and area-level factors affecting variation in SEN provision across different phenotypes (WP2). WP3 assesses the impact of SEN provision on health and education outcomes for specific health phenotypes using a range of causal inference methods to account for confounding factors and possible selection bias. In WP4 we review local policies and synthesise findings from surveys, interviews and focus groups of service users and providers to understand factors associated with variation in and experiences of identification, assessment and provision for SEN. Triangulation of findings on outcomes, variation and impact of SEN provision for different health phenotypes in ECHILD, with experiences of SEN provision will inform interpretation of findings for policy, practice and families and methods for future evaluation.

Research ethics committees have approved the use of the ECHILD database and, separately, the survey, interviews and focus groups of young people, parents and service providers. These stakeholders will contribute to the design, interpretation and communication of findings.