Computed tomography referral guidelines adherence in Europe: insights from a seven-country audit.

BACKGROUND: Ensuring appropriate computed tomography (CT) utilization optimizes patient care while minimizing radiation exposure. Decision support tools show promise for standardizing appropriateness. OBJECTIVES: In the current study, we aimed to assess CT appropriateness rates using the European Society of Radiology (ESR) iGuide criteria across seven European countries. Additional objectives were to identify factors associated with appropriateness variability and examine recommended alternative exams. METHODS: As part of the European Commission-funded EU-JUST-CT project, 6734 anonymized CT referrals were audited across 125 centers in Belgium, Denmark, Estonia, Finland, Greece, Hungary, and Slovenia. In each country, two blinded radiologists independently scored each exam's appropriateness using the ESR iGuide and noted any recommended alternatives based on presented indications. Arbitration was used in case auditors disagreed. Associations between appropriateness rate and institution type, patient's age and sex, inpatient/outpatient patient status, anatomical area, and referring physician's specialty were statistically examined within each country. RESULTS: The average appropriateness rate was 75%, ranging from 58% in Greece to 86% in Denmark. Higher rates were associated with public hospitals, inpatient settings, and referrals from specialists. Variability in appropriateness existed by country and anatomical area, patient age, and gender. Common alternative exam recommendations included magnetic resonance imaging, X-ray, and ultrasound. CONCLUSION: This multi-country evaluation found that even when using a standardized imaging guideline, significant variations in CT appropriateness persist, ranging from 58% to 86% across the participating countries. The study provided valuable insights into real-world utilization patterns and identified opportunities to optimize practices and reduce clinical and demographic disparities in CT use. KEY POINTS: Question Largest multinational study (7 EU countries, 6734 CT referrals) assessed real-world CT appropriateness using ESR iGuide, enabling cross-system comparisons. Findings Significant variability in appropriateness rates across institution type, patient status, age, gender, exam area, and physician specialty, highlighted the opportunities to optimize practices based on local factors. Clinical relevance International collaboration on imaging guidelines and decision support can maximize CT benefits while optimizing radiation exposure; ongoing research is crucial for refining evidence-based guidelines globally.

Skeletal muscle size and quality in healthy kidney donors, normal range and clinical associations.

The gold standard to estimate muscle mass and quality is computed tomography (CT) scan. Lower mass and density (intramuscular fat infiltration) of skeletal muscles are markers of sarcopenia, associated with increased mortality risk, impaired physical function, and poorer prognosis across various populations and medical conditions. We aimed to describe standard reference values in healthy population, prospective kidney donors, and correlate clinical parameters to muscle mass and density. Included in the cohort 384 consecutive kidney donors. Mean age was 44.6 ± 11.5 (range 18.4-74.2), 46% were female and mean BMI was 25.6 ± 3.8 kg/m2. Our quantified reference values for psoas cross -sectional area (CSA) index at L3 level (males/females respectively) were 6.3 ± 1.8 and 4.8 ± 1.9 cm2 /m2, and density was 46.1 ± 5 and 41 ± 5 HU at that level. Older age (standardized beta coefficient - 0.12, p = 0.04), sex (- 0.32, p < 0.001) and BMI (0.17, p = 0.002) were significantly associated with CSA index of psoas at L3. Density, however, was associated with triglycerides level (- 0.21, p < 0.001), in addition to age (- 0.22, p < 0.0001), sex (- 0.27, p < 0.001) and BMI (- 0.1, p = 0.05). Our study validates the normative values of psoas muscle mass and density in healthy individuals and suggests correlations with clinical parameters. We demonstrate the significance of measuring not only the mass of the muscle, but also its density, as it has a valid association with metabolic parameters, including BMI and lipid level, even in healthy individuals and in the normal range of the tests.

Optimizing outcomes: A retrospective analysis of discharge safety for elderly mTBI patients on anticoagulation therapy.

BACKGROUND: Mild traumatic brain injuries (mTBIs) pose a significant risk, particularly in the elderly population on anticoagulation therapy. The safety of discharging these patients from the emergency department (ED) with a negative initial computed tomography (CT) scan has been debated due to the risk of delayed intracranial hemorrhage (d-ICH). OBJECTIVE: To compare outcomes, including d-ICH, between elderly patients on anticoagulation therapy presenting with mTBI who were admitted versus discharged from the ED after an initial negative head CT scan. METHODS: We conducted a retrospective observational study at the Chaim Sheba Medical Center, assessing outcomes of 1598 elderly patients on anticoagulation therapy who presented with mTBI and an initial negative head CT scan. Patients were either admitted for 24-h observation (Group A, n = 829) or discharged immediately from the ED (Group B, n = 769). The primary outcome was incidence of d-ICH within 14 days. RESULTS: Among the 1598 patients included in the study, 46 admitted patients and 1 discharged patient returned within 14 days for repeat CT, identifying one asymptomatic hemorrhage in the discharged patient. Mortality at 30 days was significantly higher in admitted patients compared to discharged patients (4.8% vs. 1.8%, p = 0.001), though cause of death was unrelated to head injury in both groups. CONCLUSION: In elderly patients on anticoagulation with mTBI and a negative initial CT, admission was associated with a higher risk of d-ICH compared to discharge. These findings have implications for clinical decision-making in this high-risk population.

Augmenting intensive care unit nursing practice with generative AI: A formative study of diagnostic synergies using simulation-based clinical cases.

BACKGROUND: As generative artificial intelligence (GenAI) tools continue advancing, rigorous evaluations are needed to understand their capabilities relative to experienced clinicians and nurses. The aim of this study was to objectively compare the diagnostic accuracy and response formats of ICU nurses versus various GenAI models, with a qualitative interpretation of the quantitative results. METHODS: This formative study utilized four written clinical scenarios representative of real ICU patient cases to simulate diagnostic challenges. The scenarios were developed by expert nurses and underwent validation against current literature. Seventy-four ICU nurses participated in a simulation-based assessment involving four written clinical scenarios. Simultaneously, we asked ChatGPT-4 and Claude-2.0 to provide initial assessments and treatment recommendations for the same scenarios. The responses from ChatGPT-4 and Claude-2.0 were then scored by certified ICU nurses for accuracy, completeness and response. RESULTS: Nurses consistently achieved higher diagnostic accuracy than AI across open-ended scenarios, though certain models matched or exceeded human performance on standardized cases. Reaction times also diverged substantially. Qualitative response format differences emerged such as concision versus verbosity. Variations in GenAI models system performance across cases highlighted generalizability challenges. CONCLUSIONS: While GenAI demonstrated valuable skills, experienced nurses outperformed in open-ended domains requiring holistic judgement. Continued development to strengthen generalized decision-making abilities is warranted before autonomous clinical integration. Response format interfaces should consider leveraging distinct strengths. Rigorous mixed methods research involving diverse stakeholders can help iteratively inform safe, beneficial human-GenAI partnerships centred on experience-guided care augmentation. RELEVANCE TO CLINICAL PRACTICE: This mixed-methods simulation study provides formative insights into optimizing collaborative models of GenAI and nursing knowledge to support patient assessment and decision-making in intensive care. The findings can help guide development of explainable GenAI decision support tailored for critical care environments. PATIENT OR PUBLIC CONTRIBUTION: Patients or public were not involved in the design and implementation of the study or the analysis and interpretation of the data.

Comparing Emergency Nursing Measures Before and During COVID-19: A Retrospective Study of Assessment, Triage, and Workflow.

Background: The COVID-19 pandemic significantly impacted emergency department (ED) operations and patient care. Understanding its effects on nursing processes, triage accuracy, and wait times is pivotal for optimizing outcomes. Objectives: This study aimed to analyze the differences in nursing processes, triage accuracy, and wait times before and during the COVID-19 pandemic. Design: A retrospective cohort study. Methods: The study analyzed 224 electronic medical records from a single ED, with 120 records from the pre-pandemic period (January 2019-February 2020) and 104 records from the pandemic period (March 2020-March 2021). Dependent variables included missed nursing care per validated scales, triage accuracy per Emergency Severity Index, and wait times for nursing triage and physician examination. Independent factors encompassed sociodemographic, clinical characteristics, and organization dynamics. Results: Sociodemographic and clinical profiles were comparable between periods. Triage accuracy remained high except for older patients. Nursing triage wait times differed little, yet physician examination and urgent case waits decreased amidst the pandemic. Nursing documentation completeness, such as recording patient status and mental state, augmented during this crisis period. Conclusion: This evaluation identified differences in triage accuracy, wait times, and documentation completeness before and during the COVID-19 pandemic period at a single institution. Patient age and clinical status influenced some metrics. Lessons from comparing precrisis benchmarks to intra-pandemic nursing performance may guide pandemic preparedness strategies. Further research is warranted to optimize emergency processes and outcomes during public health emergencies, as well as examine strategies through multicenter investigations comparing prepandemic to intra-pandemic performance to provide broader insights into challenges and inform efforts to bolster emergency care through future crises.

Evaluating the Accuracy and Impact of the ESR-iGuide Decision Support Tool in Optimizing CT Imaging Referral Appropriateness.

Radiology referral quality impacts patient care, yet factors influencing quality are poorly understood. This study assessed the quality of computed tomography (CT) referrals, identified associated characteristics, and evaluated the ESR-iGuide clinical decision support tool's ability to optimize referrals. A retrospective review analyzed 300 consecutive CT referrals from an acute care hospital. Referral quality was evaluated on a 5-point scale by three expert reviewers (inter-rater reliability κ = 0.763-0.97). The ESR-iGuide tool provided appropriateness scores and estimated radiation exposure levels for the actual referred exams and recommended exams. Scores were compared between actual and recommended exams. Associations between ESR-iGuide scores and referral characteristics, including the specialty of the ordering physician (surgical vs. non-surgical), were explored. Of the referrals, 67.1% were rated as appropriate. The most common exams were head and abdomen/pelvis CTs. The ESR-iGuide deemed 70% of the actual referrals "usually appropriate" and found that the recommended exams had lower estimated radiation exposure compared to the actual exams. Logistic regression analysis showed that non-surgical physicians were more likely to order inappropriate exams compared to surgical physicians. Over one-third of the referrals showed suboptimal quality in the unstructured system. The ESR-iGuide clinical decision support tool identified opportunities to optimize appropriateness and reduce radiation exposure. Implementation of such a tool warrants consideration to improve communication and maximize patient care quality.

Treating wartime injuries amidst attack: insights from a medical facility on the edge of combat.

BACKGROUND: Providing emergency care during conflict poses unique challenges for frontline hospitals. Barzilai Medical Center (BUMCA) in Ashkelon, Israel is a Level I trauma center located close to the Gaza border. During the November 2023 escalation of conflict, BUMCA experienced surging numbers of civilian and military trauma patients while also coming under rocket fire. METHODS: We conducted a retrospective review of BUMCA operational records and 827 de-identified patient records from October 7-14, 2023. Records provided data on daily patient volumes, injury patterns, resource constraints, and impacts of rocket attacks on hospital function. Basic demographic data was obtained including age, gender, injury severity scores, and disposition. RESULTS: Of the 827 patients brought to BUMCA, most (n = 812, 98.2%) presented through the emergency department. Tragically, 99 individuals were pronounced dead on arrival. Injury severity assessments found nearly half (47%) had minor injuries such as lacerations, contusions and sprains, while 25% exhibited moderate injuries like deep lacerations and fractures. 15% sustained severe or critical injuries including severe head injuries. The largest age group consisted of adults aged 19-60 years. No pediatric patients were admitted despite proximity to residential neighborhoods. The majority of cases (61%) involved complex polytrauma affecting multiple body regions. BUMCA served as both the primary treatment facility and a triage hub, coordinating secondary transports to other trauma centers as needed. Patient volumes fluctuated unpredictably from 30 to an overwhelming 125 daily, straining emergency services. Resources faced shortages of beds, medical staff, supplies and disruptions to power from nearby missile impacts further challenging care delivery. CONCLUSION: Despite facing surging demand, unpredictable conditions and external threats, BUMCA demonstrated resilience in maintaining emergency trauma services through an adaptive triage approach and rapid surges in capacity. Their experience provides insights for improving frontline hospital preparedness and continuity of care during conflict through advance contingency planning and surge protocols. Analysis of patient outcomes found a mortality rate of 15% given the complex, multi-region injuries sustained by many patients. This study highlights the challenges faced and strengths exhibited by medical professionals operating under hazardous conditions in minimizing loss of life. PATIENT AND PUBLIC INVOLVEMENT IN RESEARCH: Given that the study analyzed patient data from a hospital treating casualties of an ongoing armed conflict, directly engaging patients or the public during the sensitive research process could have posed risks. The volatile security situation and restrictions and protections in place amidst the crisis made it not feasible or appropriate to involve them in the study's design, methods, reporting of results, or dissemination plans. Our aim was to conduct this retrospective analysis in a way that did not endanger those affected or compromise the hospital's emergency response operations.

Conflict-Related Breakdowns in Routine Medical Care - Longitudinal Outcomes for Civilians with Pre-Existing Cognitive Impairment.

Individuals with dementia face increased vulnerability during crises like armed conflicts. However, little is known about how conflicts affect dementia care delivery and patients' health. We conducted a longitudinal cohort study using medical record data. The study included 23,733 adults aged≥65 years with a diagnosis of dementia and 249,749 matched adults without dementia. Data were collected at baseline (March-October 2023), and two follow-up timepoints (December 2023 and February 2024), bracketing an armed conflict between Israel and Palestinian militant groups that began on October 7, 2023. We compared changes over time in clinical characteristics, medication use, healthcare utilization, costs between groups. Dementia prevalence was stable, but psychotropic medication use declined more sharply in those with dementia. Rates of depression diagnoses fell, and obesity rose in both groups. Healthcare utilization decreased substantially post-conflict, with fewer outpatient visits, hospitalizations, and emergency visits. Cost divergence between groups also increased over time. Machine learning identified shifting clusters of service users from high to mainly low users' post-conflict. The conflict severely disrupted routine dementia care and altered health behaviors. Flexible service delivery and access promotion strategies are needed to support vulnerable populations like people with dementia during crises.

Modeling mortality prediction in older adults with dementia receiving COVID-19 vaccination.

OBJECTIVE: This study compared COVID-19 outcomes between vaccinated and unvaccinated older adults with and without cognitive impairment. METHOD: Electronic health records from Israel from March 2020-February 2022 were analyzed for a large cohort (N = 85,288) aged 65 + . Machine learning constructed models to predict mortality risk from patient factors. Outcomes examined were COVID-19 mortality and hospitalization post-vaccination. RESULTS: Our study highlights the significant reduction in mortality risk among older adults with cognitive disorders following COVID-19 vaccination, showcasing a survival rate improvement to 93%. Utilizing machine learning for mortality prediction, we found the XGBoost model, enhanced with inverse probability of treatment weighting, to be the most effective, achieving an AUC-PR value of 0.89. This underscores the importance of predictive analytics in identifying high-risk individuals, emphasizing the critical role of vaccination in mitigating mortality and supporting targeted healthcare interventions. CONCLUSIONS: COVID-19 vaccination strongly reduced poor outcomes in older adults with cognitive impairment. Predictive analytics can help identify highest-risk cases requiring targeted interventions.

Harnessing innovation to help meet the needs of elders: field testing an electronic tool to streamline geriatric assessments across healthcare settings.

BACKGROUND: As populations age globally, effectively managing geriatric health poses challenges for primary care. Comprehensive geriatric assessments (CGAs) aim to address these challenges through multidisciplinary screening and coordinated care planning. However, most CGA tools and workflows have not been optimised for routine primary care delivery. OBJECTIVE: This study aimed to evaluate the impact of a computerised CGA tool, called the Golden Age Visit, implemented in primary care in Israel. METHODS: This study employed a quasiexperimental mixed-methods design to evaluate outcomes associated with the Golden Age electronic health assessment tool. Quantitative analysis used electronic medical records data from Maccabi Healthcare Services, the second largest health management organisation (HMO) in Israel. Patients aged 75 and older were included in analyses from January 2017 to December 2019 and January 2021 to December 2022. For patients, data were also collected on controls who did not participate in the Golden Age Visit programme during the same time period, to allow for comparison of outcomes. For physicians, qualitative data were collected via surveys and interviews with primary care physicians who used the Golden Age Visit SMARTEST e-assessment tool. RESULTS: A total of 9022 community-dwelling adults aged 75 and older were included in the study: 1421 patients received a Golden Age Visit CGA (intervention group), and 7601 patients did not receive the assessment (control group). After CGAs, diagnosis rates increased significantly for neuropsychiatric conditions and falls. Referrals to physiotherapy, occupational therapy, dietetics and geriatric outpatient clinics also rose substantially. However, no differences were found in rates of hip fracture or relocation to long-term care between groups. Surveys among physicians (n=151) found high satisfaction with the programme. CONCLUSION: Implementation of a large-scale primary care CGA programme was associated with improved diagnosis and management of geriatric conditions. Physicians were also satisfied, suggesting good uptake and feasibility within usual care. Further high-quality studies are still needed but these results provide real-world support for proactively addressing geriatric health needs through structured screening models.

An evaluation of the capabilities of language models and nurses in providing neonatal clinical decision support.

AIM: To assess the clinical reasoning capabilities of two large language models, ChatGPT-4 and Claude-2.0, compared to those of neonatal nurses during neonatal care scenarios. DESIGN: A cross-sectional study with a comparative evaluation using a survey instrument that included six neonatal intensive care unit clinical scenarios. PARTICIPANTS: 32 neonatal intensive care nurses with 5-10 years of experience working in the neonatal intensive care units of three medical centers. METHODS: Participants responded to 6 written clinical scenarios. Simultaneously, we asked ChatGPT-4 and Claude-2.0 to provide initial assessments and treatment recommendations for the same scenarios. The responses from ChatGPT-4 and Claude-2.0 were then scored by certified neonatal nurse practitioners for accuracy, completeness, and response time. RESULTS: Both models demonstrated capabilities in clinical reasoning for neonatal care, with Claude-2.0 significantly outperforming ChatGPT-4 in clinical accuracy and speed. However, limitations were identified across the cases in diagnostic precision, treatment specificity, and response lag. CONCLUSIONS: While showing promise, current limitations reinforce the need for deep refinement before ChatGPT-4 and Claude-2.0 can be considered for integration into clinical practice. Additional validation of these tools is important to safely leverage this Artificial Intelligence technology for enhancing clinical decision-making. IMPACT: The study provides an understanding of the reasoning accuracy of new Artificial Intelligence models in neonatal clinical care. The current accuracy gaps of ChatGPT-4 and Claude-2.0 need to be addressed prior to clinical usage.

Evaluating emergency response at a hospital near the Gaza border within 24 h of increased conflict.

BACKGROUND: Frontline hospitals near active hostilities face unique challenges in delivering emergency care amid threats to infrastructure and personnel safety. Existing literature focuses on individual aspects like mass casualty protocols or medical neutrality, with limited analysis of operating acute services directly under fire. OBJECTIVES: To describe the experience of a hospital situated meters from hostilities and analyze strategies implemented for triage, expanding surge capacity, and maintaining continuity of care during attacks with limited medical staff availability due to hazardous conditions. A focus will be placed on assessing how the hospital functioned and adapted care delivery models in the event of staffing limitations preventing all teams from arriving on site. METHODS: A retrospective case study was conducted of patient records from Barzilai University Medical Center at Ashkelon (BUMCA) Medical Center in Israel within the first 24 h after escalated conflict began on October 7, 2023. Data on 232 admissions were analyzed regarding demographics, treatment protocols, time to disposition, and mortality. Missile alert data correlated patient surges to attacks. Statistical and geospatial analyses were performed. RESULTS: Patients predominantly male soldiers exhibited blast/multisystem trauma. Patient surges at the hospital were found to be correlated with the detection of incoming missile attacks from Gaza within 60 min of launch. While 131 (56%) patients were discharged and 55 (24%) transferred within 24 h, probabilities of survival declined over time reflecting injury severity limitations. 31 deaths occurred from severe presentation. CONCLUSION: Insights gleaned provide a compelling case study on managing mass casualties at the true frontlines. By disseminating BUMCA's trauma response experience, strategies can strengthen frontline hospital protocols optimizing emergency care delivery during hazardous armed conflicts through dynamic surge capacity expansion, early intervention prioritization, and infrastructure/personnel protection measures informed by risks.

Optimising wound monitoring: Can digital tools improve healing outcomes and clinic efficiency.

BACKGROUND: Chronic wounds present significant challenges for patients and nursing care teams worldwide. Digital health tools offer potential for more standardised and efficient nursing care pathways but require further rigorous evaluation. OBJECTIVE: This retrospective matched cohort study aimed to compare the impacts of a digital tracking application for wound documentation versus traditional manual nursing assessments. METHODS: Data from 5236 patients with various wound types were analysed. Propensity score matching balanced groups, and bivariate tests, correlation analyses, linear regression, and Hayes' Process Macro Model 15 were utilised for a mediation-moderation model. RESULTS: Digital wound tracking was associated with significantly shorter healing durations (15 vs. 35 days) and fewer clinic nursing visits (3 vs. 5.8 visits) compared to standard nursing monitoring. Digital tracking demonstrated improved wound size reduction over time. Laboratory values tested did not consistently predict healing outcomes. Digital tracking exhibited moderate negative correlations with the total number of nursing visits. Regression analysis identified wound complexity, hospitalizations, and initial wound size as clinical predictors for more nursing visits in patients with diabetes mellitus (p < .01). Digital tracking significantly reduced the number of associated nursing visits for patients with peripheral vascular disease. CONCLUSION: These findings suggest that digital wound management may streamline nursing care and provide advantages, particularly for comorbid populations facing treatment burdens. REPORTING METHOD: This study adhered to STROBE guidelines in reporting this observational research. RELEVANCE TO CLINICAL PRACTICE: By streamlining documentation and potentially shortening healing times, digital wound tracking could help optimise nursing resources, enhance wound care standards, and improve patient experiences. This supports further exploration of digital health innovations to advance evidence-based nursing practice. PATIENT OR PUBLIC CONTRIBUTION: This study involved retrospective analysis of existing patient records and did not directly include patients or the public in the design, conduct, or reporting of the research.

Exploring the Self-Reported Physical and Psychological Effects in a Population Exposed to a Regional Conflict.

BACKGROUND: Conflict profoundly impacts community health and well-being. While post-conflict research exists, little is known about initial effects during active hostilities. OBJECTIVE: To assess self-reported changes in health behaviors, distress, and care access within one month of regional warfare onset in a conflict-affected community. METHODS: An online survey was conducted in November 2023 among 501 residents (mean age 40.5 years) of a community where war began October 7th. Measures evaluated physical health, mental health, diet, substance use, sleep, weight changes, and healthcare access before and after the declaration of war. RESULTS: Relative to pre-war, respondents reported significantly increased rates of tobacco (56%) and alcohol (15%) consumption, worsening sleep quality (63%), elevated distress (18% sought help; 14% needed but didn't receive it), and postponed medical care (36%). Over a third reported weight changes. Distress was higher among females and those endorsing maladaptive coping. CONCLUSION: Within one month, substantial impacts on community psychosocial and behavioral health emerged. Unmet mental health needs and risk-taking behaviors were early indicators of conflict's health consequences. Continuous monitoring of conflict-affected communities is needed to inform tailored interventions promoting resilience and prevent entrenchment of harms over time.

A comparative vignette study: Evaluating the potential role of a generative AI model in enhancing clinical decision-making in nursing.

AIM: This study explores the potential of a generative artificial intelligence tool (ChatGPT) as clinical support for nurses. Specifically, we aim to assess whether ChatGPT can demonstrate clinical decision-making equivalent to that of expert nurses and novice nursing students. This will be evaluated by comparing ChatGPT responses to clinical scenarios to those of nurses on different levels of experience. DESIGN: This is a cross-sectional study. METHODS: Emergency room registered nurses (i.e. experts; n = 30) and nursing students (i.e. novices; n = 38) were recruited during March-April 2023. Clinical decision-making was measured using three validated clinical scenarios involving an initial assessment and reevaluation. Clinical decision-making aspects assessed were the accuracy of initial assessments, the appropriateness of recommended tests and resource use and the capacity to reevaluate decisions. Performance was also compared by timing response generations and word counts. Expert nurses and novice students completed online questionnaires (via Qualtrics), while ChatGPT responses were obtained from OpenAI. RESULTS: Concerning aspects of clinical decision-making and compared to novices and experts: (1) ChatGPT exhibited indecisiveness in initial assessments; (2) ChatGPT tended to suggest unnecessary diagnostic tests; (3) When new information required re-evaluation, ChatGPT responses demonstrated inaccurate understanding and inappropriate modifications. In terms of performance, the mean number of words utilized in ChatGPT answers was 27-41 times greater than that utilized by both experts and novices; and responses were provided approximately 4 times faster than those of novices and twice faster than expert nurses. ChatGPT responses maintained logical structure and clarity. CONCLUSIONS: A generative AI tool demonstrated indecisiveness and a tendency towards over-triage compared to human clinicians. IMPACT: The study shows that it is important to approach the implementation of ChatGPT as a nurse's digital assistant with caution. More study is needed to optimize the model's training and algorithms to provide accurate healthcare support that aids clinical decision-making. REPORTING METHOD: This study adhered to relevant EQUATOR guidelines for reporting observational studies. PATIENT OR PUBLIC CONTRIBUTION: Patients were not directly involved in the conduct of this study.

Identifying Diabetes Related-Complications in a Real-World Free-Text Electronic Medical Records in Hebrew Using Natural Language Processing Techniques.

BACKGROUND: Studies have demonstrated that 50% to 80% of patients do not receive an International Classification of Diseases (ICD) code assigned to their medical encounter or condition. For these patients, their clinical information is mostly recorded as unstructured free-text narrative data in the medical record without standardized coding or extraction of structured data elements. Leumit Health Services (LHS) in collaboration with the Israeli Ministry of Health (MoH) conducted this study using electronic medical records (EMRs) to systematically extract meaningful clinical information about people with diabetes from the unstructured free-text notes. OBJECTIVES: To develop and validate natural language processing (NLP) algorithms to identify diabetes-related complications in the free-text medical records of patients who have LHS membership. METHODS: The study data included 2.3 million records of 41 469 patients with diabetes aged 35 or older between the years 2012 and 2017. The diabetes related complications included cardiovascular disease, diabetic neuropathy, nephropathy, retinopathy, diabetic foot, cognitive impairments, mood disorders and hypoglycemia. A vocabulary list of terms was determined and adjudicated by two physicians who are experienced in diabetes care board certified diabetes specialist in endocrinology or family medicine. Two independent registered nurses with PhDs reviewed the free-text medical records. Both rule-based and machine learning techniques were used for the NLP algorithm development. Precision, recall, and F-score were calculated to compare the performance of (1) the NLP algorithm with the reviewers' comments and (2) the ICD codes with the reviewers' comments for each complication. RESULTS: The NLP algorithm versus the reviewers (gold standard) achieved an overall good performance with a mean F-score of 86%. This was better than the ICD codes which achieved a mean F-score of only 51%. CONCLUSION: NLP algorithms and machine learning processes may enable more accurate identification of diabetes complications in EMR data.

Priority setting for health technology adoption at the national level: Lessons learned over 25 years' experience.

BACKGROUND: Limited health budgets and continual advancement of health technologies require mechanisms for prioritization. Israel, with a publicly funded health service basket, has implemented and optimized such a health technology assessment process since 1999.We describe the process of evaluating technologies according to the Israeli model, analyze its outputs and benefits over two decades of implementation, and compare its key features with international experience. METHODS: Retrospective data were collected between 1998 and 2023, including work processes, committee composition, number of applications submitted and approved by a clinical domain, and yearly cost of the basket. Features were evaluated within the evidence-informed deliberative process (EDP) framework. RESULTS: This national model involves relevant stake holders in a participatory and transparent process, in a timely manner, and is accepted by the public, health professionals, and policy makers, facilitating early adoption of the newest medical technologies. Between 11 and 19 percent of applications are approved for reimbursement annually, mostly pharmaceuticals. On average 26 percent of approved technologies are added to the list without additional budget. Major domains of approved technologies were oncology, cardiology, and neurology. CONCLUSIONS: Israel created a unique model for the expansion of the health service basket. Despite an increasing number of applications and rising costs, the mechanism enables a consensus to be reached on which technologies to fund, while remaining within budget constraints and facilitating immediate implementation. The process, which prioritizes transparency and stake holder involvement, allows just a resource allocation while maximizing the adoption of novel technologies, contributing to an outstanding national level of health despite relatively low health spending.