UNASSIGNED: Given the strikingly high diagnostic error rate in hospitals, and the recent development of Large Language Models (LLMs), we set out to measure the diagnostic sensitivity of two popular LLMs: GPT-4 and PaLM2. Small-scale studies to evaluate the diagnostic ability of LLMs have shown promising results, with GPT-4 demonstrating high accuracy in diagnosing test cases. However, larger evaluations on real electronic patient data are needed to provide more reliable estimates.
UNASSIGNED: To fill this gap in the literature, we used a deidentified Electronic Health Record (EHR) data set of about 300,000 patients admitted to the Beth Israel Deaconess Medical Center in Boston. This data set contained blood, imaging, microbiology and vital sign information as well as the patients\' medical diagnostic codes. Based on the available EHR data, doctors curated a set of diagnoses for each patient, which we will refer to as ground truth diagnoses. We then designed carefully-written prompts to get patient diagnostic predictions from the LLMs and compared this to the ground truth diagnoses in a random sample of 1000 patients.
UNASSIGNED: Based on the proportion of correctly predicted ground truth diagnoses, we estimated the diagnostic hit rate of GPT-4 to be 93.9%. PaLM2 achieved 84.7% on the same data set. On these 1000 randomly selected EHRs, GPT-4 correctly identified 1116 unique diagnoses.
UNASSIGNED: The results suggest that artificial intelligence (AI) has the potential when working alongside clinicians to reduce cognitive errors which lead to hundreds of thousands of misdiagnoses every year. However, human oversight of AI remains essential: LLMs cannot replace clinicians, especially when it comes to human understanding and empathy. Furthermore, a significant number of challenges in incorporating AI into health care exist, including ethical, liability and regulatory barriers.

BACKGROUND: Advanced therapies offer unprecedented opportunities for treating rare neurological disorders (RNDs) in children. However, health literacy, perceptions and understanding of novel therapies need elucidation across the RND community. This study explored healthcare professionals\' and carers\' perspectives of advanced therapies in childhood-onset RNDs.
METHODS: In this mixed-methodology cross-sectional study, 20 healthcare professionals (clinicians, genetic counsellors and scientists) and 20 carers completed qualitative semistructured interviews and custom-designed surveys. Carers undertook validated psychosocial questionnaires. Thematic and quantitative data analysis followed.
RESULTS: Participants described high positive interest in advanced therapies, but low knowledge of, and access to, reliable information. The substantial \'therapeutic gap\' and \'therapeutic odyssey\' common to RNDs were recognised in five key themes: (i) unmet need and urgency for access; (ii) seeking information; (iii) access, equity and sustainability; (iv) a multidisciplinary and integrated approach to care and support and (v) difficult decision-making. Participants were motivated to intensify RND clinical trial activity and access to advanced therapies; however, concerns around informed consent, first-in-human trials and clinical trial procedures were evident. There was high-risk tolerance despite substantial uncertainties and knowledge gaps. RNDs with high mortality, increased functional burdens and no alternative therapies were consistently prioritised for the development of advanced therapies. However, little consensus existed on prioritisation to treatment access.
CONCLUSIONS: This study highlights the need to increase clinician and health system readiness for the clinical translation of advanced therapeutics for RNDs. Co-development and use of educational and psychosocial resources to support clinical decision-making, set therapeutic expectations and promotion of equitable, effective and safe delivery of advanced therapies are essential.
UNASSIGNED: Participant insights into the psychosocial burden and information need to enhance the delivery of care in this formative study are informing ongoing partnerships with families, including co-production and dissemination of psychoeducational resources featuring their voices hosted on the Sydney Children\'s Hospitals Network website SCHN Brain-Aid Resources.

This scoping review aimed to determine whether the COVID-19 pandemic influenced any modifications to patient selection methods or prioritisation and services provided by proton therapy (PT) centres. This review was conducted based on the PRISMA methodology and Joanna Briggs Institute scoping review guidelines. A literature search was performed in Medline, Embase, Web of Science and Scopus, as well as grey literature. Keywords such as \"COVID-19\" and \"Proton Therapy\" were used. Articles published from 1 January 2020 in English were included. In total, 138 studies were identified of which 11 articles met the inclusion criteria. A scoping review design was chosen to capture the full extent of information published relating to the aim. Six of 11 articles included statements regarding treatment of COVID-19 patients. Three publications recommended deferred or alternative treatment, two indicated to treat urgent/emergency patients and one reported continuous treatment for infectious patients. Recurring impacts on PT provision included more frequent use of unconventional therapies, reduced referrals, delayed treatment starts and CT simulation, change in treatment target volumes and staffing limitations due to pandemic restrictions. Consequently, telehealth consults, remote work, reduction in patient visitors, screening procedures and rigorous cleaning protocols were recommended. Few publications detailed changes to patient selection or workflow methods during the pandemic. Further research is needed to obtain more detailed information regarding current global patient selection methods in PT, collecting this data could aid in future planning for PT in Australia.

OBJECTIVE: The aim of this study was to establish the current level of collaboration between nurses and medical doctors (MDs) in the making of clinical decisions.
METHODS: Descriptive qualitative design was applied in this study.
METHODS: Semi-structured interviews were conducted to collect qualitative data. Contents were arranged according to their similarities, whereas content analysis was used to identify explanatory themes.
RESULTS: Nurses feel disrespected when medical doctors (MDs) ignore their opinions. The impression of lower level of education of nurses is seen as a cause to their opinions been ignored by the medical doctors. Nurses sometimes adhere to the instructions of MDs, but on other times, they carry on with their own proposed treatment.
UNASSIGNED: Involvement of nurses in clinical decisions will enable nurses to effectively advocate for patients.

OBJECTIVE: To describe ambulance nurses\' experience of deciding a patient does not require ambulance care.
METHODS: An inductive, empirical study with a qualitative approach.
METHODS: Data collection was conducted through semi-structured interviews, and collected data were analysed with qualitative manifest content analysis. Data were collected during the spring 2017, and eight ambulance nurses participated.
RESULTS: The findings are presented in one main category, which is \"Not very ill but a difficult decision\" with totally three subcategories. The ambulance nurse\'s experience of making the assessment when the patient has no need for ambulance care is like walking the balance of slack line. This means that the assessment can be both easy and very difficult but something that definitely requires experience, knowledge and dedication.

OBJECTIVE: To explore the deliberation and enactment processes of nurses in relation to pain and other discomforts in the critically ill patients after the implementation of an analgosedation protocol.
BACKGROUND: Nurses in intensive care units (ICU) face great challenges when managing pain and other discomforts and distinguishing between patients\' needs for analgesics and sedatives. An analgosedation protocol favouring pain management, light sedation and early mobilization was implemented in a university hospital ICU in Norway in 2014. Changing sedation paradigms resulting in an increasing number of awake patients during critical illness is expected to affect nursing practice.
METHODS: Exploratory, single-unit study in a mixed adult ICU.
METHODS: Data collection with participant observation and semi-structured interviews in sixteen clinical situations in 2014 and 2015. Thirteen experienced certified critical care nurses were included. Thematic content analysis was conducted.
RESULTS: An overall theme \"Having the compass-drawing the map\" emerged from the analysis. The protocol or strategy of analgosedation appeared to provide a direction for treatment and care, although requiring extensive interpretation of needs and individualization of care, often in challenging situations. The overall theme was abstracted from three themes: \"Interpreting a complex whole,\" \"Balancing conflicting goals\" and \"Experiencing strain from acting across ideals.\"
CONCLUSIONS: Nurses seem to attend adequately to patient pain, but the approach to discomforts other than pain appears unsystematic and haphazard. More explicit goals of care and strategies to handle discomfort as distinct from pain are needed. More research is needed to identify effective comfort measures for ICU patients.

In Japan, the do not attempt resuscitation (DNAR) order is practised routinely even though no related laws or guidelines exist. This study aimed to clarify the current status of DNAR, reveal existing DNAR-related issues, and improve the application of DNAR.
A questionnaire survey of medical institutions in Kanagawa Prefecture (total population, 9,120,000) about the current status of DNAR was carried out.
The results showed that DNAR has been practised at approximately 90% of the hospitals surveyed, but only about 30% have developed in-hospital DNAR guidelines. Approximately 80% of the hospitals do not involve patients in the decision on their own DNAR orders. Because the DNAR order has not been legislated, it is often unclear whether to resuscitate patients when a request for an ambulance is made for a cardiac arrest at home.
It is necessary for prefectures, municipalities, and local medical control authorities to take the initiative in establishing an ordinance on DNAR orders and developing guidelines.