Risk Assessment Tool Research Articles

Cognitive Risk Stratification Score in Middle-aged and Older Adults with Type 2 Diabetes: a Cross-Sectional Study.

Cognitive impairment (CI) affects approximately 45% of middle-aged and older adults with Type 2 Diabetes Mellitus (T2DM) globally. Although formal comprehensive neuropsychological tests are the gold standard for diagnosing CI, they are often time-intensive and may not be feasible in primary care. This study aimed to develop and validate a novel Risk Stratification Score (RSS) to rapidly and comprehensively predict CI risk among middle-aged and older adults with T2DM, offering a streamlined alternative in clinical practice. A cross-sectional study was conducted from July 2023 to February 2024 in a primary care polyclinic in Singapore's western region. Participants aged between 40 and 85 diagnosed with T2DM (n=150) were included in a convenience sampling. The primary outcome was CI status and assessed using formal neuropsychological tests, which including Montreal Cognitive Assessment (MoCA). CI was identified in 49.3% of participants (n=74). The RSS, incorporating the MoCA, Diastolic Blood Pressure (DBP), and Short Physical Performance Battery (SPPB), demonstrated excellent discrimination, achieving an area under the ROC curve of 0.802 (p < 0.001). With an optimal cutoff of 0.3, the model showed a sensitivity of 63.5% and specificity of 86.8%, effectively differentiating high- and low-risk CI groups. RSS in clinical practice, exemplified by the Integrated Metabolic Cognitive Risk Stratification Pathway (imCRSP), is a promising tool for rapid CI risk assessment in primary care. Its robust predictive accuracy and ease of use support its application for early intervention in middle-aged and older adults with T2DM. Future studies should validate its use longitudinally and across diverse populations to enhance generalizability.

Effects of open-skill exercise on executive functions in children and adolescents: a systematic review and meta-analysis

BackgroundThe research on the intervention of open-skill exercise on the executive functions of children and adolescents still requires quantitative synthesis, and there is inconsistency in the effects of intervention by strategic and interceptive skills, which are sub-divided from open-skill exercise. Therefore, this study systematically explores the aforementioned issues and examines the potential moderating factors in the effects of open-skill exercise intervention on executive functions.MethodsComputer searches of the CNKI, WOS, PubMed, ScienceDirect and SPORTDiscus databases were conducted. Two researchers independently screened the articles and extracted data, and used the bias risk assessment tool recommended by the Cochrane Collaboration Network and the Methodological Index for Non- Randomized Studies (MINORS) scale to assess included randomized controlled trials (RCTs) and quasi-experimental designs (QEDs). Statistical analyses were performed using Stata 16.0 software.ResultsA total of 16 articles and 17 studies were included, comprising 11 RCTs and 6 QEDs. The participants were 1,298 children and adolescents aged 5 to 16. Open skill exercises have significant intervention effects (p &amp;lt; 0.01) on inhibitory control (SMD = −0.627, 95%CI = −0.890 to −0.363), working memory (SMD = −0.517, 95%CI = −0.768 to −0.266), and cognitive flexibility (SMD = −0.652, 95%CI = −1.085 to −0.219). The effects of strategic skill exercises are higher than those of interceptive skills, particularly in the dimension of inhibitory control (SMD = −0.707, 95%CI = −0.819 to −0.594, p &amp;lt; 0.05). In addition, moderate-intensity and higher-frequency exercises overall have a more positive effect on promoting executive functions (p &amp;lt; 0.05); interventions of 6 to 10 weeks are more effective for working memory (p &amp;lt; 0.05), while 30-min sessions are the most effective for working memory (p &amp;lt; 0.05), and sessions lasting 75 to 120 min are the most effective for cognitive flexibility (p &amp;lt; 0.05). Open-skill exercise has a more positive impact on inhibitory control in the 5–9 age group and on working memory in the 10–16 age group (p &amp;lt; 0.05); open-skill exercise, especially, has a more positive intervention effect on inhibitory control in the Eastern group (p &amp;lt; 0.05). Both Egger linear regression analyses and literature sensitivity analyses suggested that the Meta-analysis results were stable and reliable.ConclusionOpen-skill exercise has a positive intervention effect on executive functioning in children and adolescents, and strategic skill exercise interventions are more effective. In addition, the quantitative elements of exercise (intensity, frequency, and duration per session) and demographic factors (age and ethnicity) play a potential moderating role in this context. Based on this, it is recommended that children and adolescents choose more strategic open-skill exercises according to their actual situation and select appropriate quantitative exercise factors to maximize the enhancement of their executive functions.Systematic review registrationhttps://www.crd.york.ac.uk/prospero/#myprospero, CRD42025636714.

Misconceptions about intimate partner violence risk assessment algorithm in the Basque Country: a reply to Valdivia, Hyde-Vaamonde, and García-Marcos (2024)

Abstract Violence risk assessment is an internationally recognised methodology, aimed to manage different forms of violence. Most risk assessment tools, as is the case of the reviewed one, are designed to protect victims in the context of pressure, little time, or little information. This paper presents a reply to Valdivia et al. (AI &amp; Society, July 2024) criticism of the algorithm for intimate partner violence risk assessment—EPV—used in the Basque Country. They concluded that more than 50% of high-risk victims are in danger, using results from a pilot version of the instrument, not the reviewed one published in 2010, nor the system in use since May 2013. In addition, qualitative information from a single professional generates global criticisms of the tool. Neither the current cut-off points nor the real weighting of the items nor the real risk management procedure are considered, and the personal opinion of a judge is assumed to be better than the use of tools when the accumulated research shows the opposite. When EPV risk assessment reports are used in courts, they may only temporarily affect some perpetrator rights, imposing restraining orders or, in the worst case, pretrial prison waiting for sentence. However, risk management can save the life of the victim. Cautions and suggestions related to the judicial context, such as improving risk reports or training judicial professionals, are shared. However, Valdivia et al.’s paper leads to misconceptions that extend to different sectors when echoing the wrong conclusions of their paper.

The Present State and Potential Applications of Artificial Intelligence in Cancer Diagnosis and Treatment.

An aberrant increase in cancer incidences has demanded extreme attention globally despite advancements in diagnostic and management strategies. The high mortality rate is concerning, and tumour heterogeneity at the genetic, phenotypic, and pathological levels exacerbates the problem. In this context, lack of early diagnostic techniques and therapeutic resistance to drugs, sole awareness among the public, coupled with the unavailability of these modern technologies in developing and low-income countries, negatively impact cancer management. One of the prime necessities of the world today is the enhancement of early detection of cancers. Several independent studies have shown that screening individuals for cancer can improve patient survival but are bogged down by risk classification and major problems in patient selection. Artificial intelligence (AI) has significantly advanced the field of oncology, addressing various medical challenges, particularly in cancer management. Leveraging extensive medical datasets and innovative computational technologies, AI, especially through deep learning (DL), has found applications across multiple facets of oncology research. These applications range from early cancer detection, diagnosis, classification, and grading, molecular characterization of tumours, prediction of patient outcomes and treatment responses, personalized treatment, and novel anti-cancer drug discovery. Over the past decade, AI/ML has emerged as a valuable tool in cancer prognosis, risk assessment, and treatment selection for cancer patients. Several patents have been and are being filed and granted. Some of those inventions were explored and are being explored in clinical settings as well. In this review, we will discuss the current status, recent advancements, clinical trials, challenges, and opportunities associated with AI/ML applications in cancer detection and management. We are optimistic about the potential of AI/ML in improving outcomes for cancer and the need for further research and development in this field.

Predicting the risk of invasive fungal infections in ICU sepsis population: the AMI risk assessment tool.

Invasive fungal infections (IFI) represent a significant contributor to mortality among sepsis patients in the Intensive Care Unit (ICU). Early diagnosis of IFI is challenging, and currently, there are no predictive tools for identifying sepsis patients who may develop IFI. Our study aims to develop a predictive scoring system to assess the risk of IFI in patients with sepsis admitted to the ICU. A retrospective collection of data from a total of 549 patients was conducted. Data-driven, clinically knowledge-driven, and decision tree models were used to identify predictive variables for risk of IFI in ICU patients with sepsis. Demographic data, vital signs, laboratory values, comorbidities, medication use, and clinical outcomes were all collected. The optimal model was selected based on model performance and clinical utility to establish a risk score. Among adult patients with sepsis admitted to the ICU, 127 patients (23.1%) developed IFI. The final data-driven model included four predictive factors, the clinically knowledge-driven model included three predictive factors, and the decision tree model included two. Based on the good performance and clinical utility of the clinically knowledge-driven model, it was chosen as the optimal risk scoring model (C-statistics: 0.79 (95% confidence interval (CI): 0.75-0.83); Hosmer-Lemeshow (H-L) test P = 0.884). The ICU sepsis patient invasive fungal infection risk (AMI) score, created based on the clinically knowledge-driven model, includes mechanical ventilation, application of immunosuppressants, and the types of antibiotics used. The C-statistics for this risk score was 0.79 (95% CI:0.75-0.84) with good calibration (H-L test P = 0.992 and see calibration curve: Fig.2). Moreover, in terms of clinical utility, the decision curve analysis for AMI showed a favorable net benefit. The application of the AMI score can effectively distinguish whether ICU sepsis patients will develop IFI, which is beneficial for clinicians to formulate targeted and timely preventive and treatment measures based on the risk of IFI.

Building resilient urban drainage systems by integrated flood risk index for evidence-based planning.

Urban flooding poses a significant risk to cities worldwide, exacerbated by increasing urbanization and climate change. Effective flood risk management requires comprehensive assessments considering the complex interaction of social, economic, and environmental factors. This study developed an innovative Urban Flood Risk Index (FRI) to quantify and assess flood risk at the sub-catchment level, providing a tool for evidence-based planning and resilient infrastructure development. This study integrates Geographic Information System (GIS), Storm Water Management Model (SWMM), Analytic Hierarchy Process (AHP), and the Pressure-State-Response (PSR) framework. The FRI incorporates seven pressure and state indicators and three response indicators weighted by expert judgment. The FRI was calculated by combining the weighted sub-indices, classifying flood risk into five levels. Results showed that 51% of the study area experienced high pressure, with 26% facing very-high pressure. The state index indicated that 55% of the area falls under a moderate state, while 21% exhibits a high state. Importantly, the response index highlighted the effectiveness of Low Impact Development (LID) practices, with 20% of the area showing high to very-high response levels. The integrated FRI demonstrated an overall moderate flood risk level for maximum sub-catchments, emphasizing the positive impact of LID practices in mitigating flood risk despite existing pressures and system limitations. This evidence-based assessment provides a valuable tool for sub-catchment level flood risk assessment. It empowers decision-makers to prioritize investments, target interventions, and develop adaptive strategies to enhance urban resilience in a changing climate.

Predictive Performance of the KINDER 1 Fall Risk Assessment Tool in a Regional Health System.

Patient fall risk assessment in the emergency department poses a unique challenge as fall risk is often associated with risk factors other than inpatient falls. While there are many inpatient fall risk assessment tools, few have been used or validated in the ED environment. Therefore, this study examined the predictive performance of the KINDER 1 Fall Risk Assessment Tool in 10 emergency departments. A retrospective cohort design was used. Data were collected from November 15, 2023, to April 30, 2024, as a part of an electronic pilot of the KINDER 1 Fall Risk Assessment tool. Inclusion criteria encompassed all adult (≥18 years) ED visits during which a KINDER 1 fall risk assessment was completed. Descriptive statistics were used to describe overall sample characteristics. Predictive performance was calculated via multiple accuracy measurements. KINDER 1 assessments were completed on 64,811 patients, of which 40 patient falls met inclusion criteria for final analysis. The mean age of the patients who fell was 58.46 years (±18.38). Final sensitivity was 77.5%, and the specificity was 75.8%. Fall prevalence was 0.06%. KINDER 1 exhibited a sufficiently high degree of sensitivity and specificity, supporting an acceptable level of predictive performance. Additional research is recommended to compare the reliability and predictive validity of KINDER 1 to the emergency Hester Davis Scale and the Memorial Emergency Department Fall Risk Assessment Tool, as well as to compare the usability of the tools for nurses in a triage setting.

Technology-Based Interventions for Pain in Children Undergoing Surgery: A Systematic Review and Meta-Analysis.

This systematic review and meta-analysis aimed to evaluate the effect of technology-based interventions on the pain of paediatric surgery patients. Recently, the number of technology-based interventions involving multimodal nonpharmacological methods tailored to pain management in paediatric surgery patients has increased. It is crucial to determine the effectiveness of these interventions. A systematic review and meta-analysis of randomized controlled trials following Cochrane methods was conducted. We performed a literature search in the Web of Science, PubMed, CINAHL, Science Direct, MEDLINE, ProQuest and Cochrane Library databases. Two independent researchers screened the literature using specific keywords and selected randomized controlled trials based on the inclusion and exclusion criteria. Each researcher extracted data and assessed the risk of bias in the randomized controlled trials using the Cochrane bias risk assessment tool. We conducted a meta-analysis on 14 randomized controlled trials included in the study. The results showed that technology-based interventions reduced pain scores in paediatric surgery patients. Cochran's Q test results pointed to a high level of heterogeneity among the randomized controlled trials. A meta-analysis result of 14 randomized controlled trials showed that technology-based interventions are effective methods for reducing pain in paediatric surgery patients. CRD42021226666.

A novel skin temperature estimation system for predicting pressure injury occurrence based on continuous body sensor data: A pilot study.

Pressure injury prevention is important in older patients with immobility. This requires an accurate and efficient prediction of the development of pressure injuries. We aimed to develop a method for estimating skin temperature changes due to ischemia and inflammation using temperature sensors placed under bedsheets to provide an objective, non-invasive, and non-constrained risk assessment tool. This study consisted of a thermal skin simulation study and a descriptive correlation study in healthy participants. A thermal skin simulation study was conducted using a model reproducing the body surface (underwear, diaper, or wet diaper conditions) and bed environment. In a descriptive-correlational study, the participants lay supine on a mattress with a temperature sensor attached to their sacral skin. The thermal skin simulation study showed that temperature changes in the skin can be estimated under the sheets by inputting time-shifted temperature data into machine learning (R2=0.9967 for underwear, 0.9950 for diapers, and 0.9869 for wet diapers). It was also demonstrated that the absolute skin temperature of a healthy individual (N=17) could be estimated with the best accuracy by inputting time-shifted data into an extra-tree regressor (R2=0.8145). A combination of interface pressure and temperature sensors can be used to estimate skin temperature changes. These findings contribute to the development of a skin temperature measurement method that can capture temperature changes over time in clinical settings.

Reducing Hospital-Acquired Pressure Injuries in a Cardiothoracic Intensive Care Unit.

Hospital-acquired pressure injuries are a significant patient safety concern. The Centers for Medicare & Medicaid Services tracks hospital-acquired pressure injuries as a patient safety indicator. Health care organizations with higher-than-expected rates may incur penalties. The pressure injury prevalence rate in a cardiothoracic intensive care unit was above the National Database of Nursing Quality Indicators benchmark. The current standard of care-use of the Braden scale for pressure injury risk assessment and the SKIN (surface, keep turning, incontinence care, and nutrition) care bundle-may not adequately address the needs of the intensive care unit population. In addition, cardiac patients present a special challenge because of their disease process and the mechanical support devices used to treat patients in cardiogenic shock, which place them at risk for the development of hospital-acquired pressure injuries. A performance improvement project was carried out in the cardiothoracic intensive care unit to reduce the prevalence and incidence of hospital-acquired pressure injuries. A preintervention convenience cohort was compared with a postintervention cohort. The intervention consisted of use of the Cubbin-Jackson scale, an intensive care unit-specific risk-assessment tool, with linked interventions to prevent pressure injuries. The preintervention and postintervention cohorts consisted of 102 patients each. The pressure injury prevalence and incidence rates decreased by 67.84% and 36.43%, respectively, from before to after the intervention. The use of an intensive care unit-specific risk-assessment tool with linked interventions to prevent pressure injury can help reduce hospital-acquired pressure injuries in an intensive care unit.

Novel Breast Cancer Risk Assessment Tools for Pre- and Post-Menopausal Asian Women: Development and Validation in a Nationwide Mammographic Screening Cohort.

Widely used breast cancer risk-prediction tools are based on data from Western countries, but risk factors may differ for Asian women. Hence, we aimed to develop a risk assessment tool for breast cancer in Asian women using a nationwide, population-based mammographic screening cohort. Women aged ≥40 years who underwent breast cancer screening and general health examination in 2009 were included. Age, body mass index (BMI), breast density, lifestyle and reproductive factors, and comorbidities were used to develop 5-year breast cancer risk-prediction models for premenopausal (n=771,856) and postmenopausal (n=1,108,047) women at baseline. The best-fit risk prediction model was constructed using backward stepwise selection in a Cox proportional hazards model and was transformed into a risk score nomogram. The performance was assessed by discrimination and calibration. In premenopausal women, high BMI, low parity, short breastfeeding period, early age at menarche, high breast density, a history of benign breast masses, and family history of breast cancer contributed to the risk prediction of breast cancer. In postmenopausal women, age, diabetes mellitus, dyslipidemia, late-onset menopause, and hormone replacement therapy use were additional risk predictors of breast cancer. Our risk-prediction model showed a concordant statistic of 0.58 (0.57-0.59) for premenopausal women and 0.64 (0.63-0.65) for postmenopausal women. The calibration plot demonstrated good correlations for both models. Our breast cancer risk-prediction model demonstrated performance comparable to that of Western countries, especially among postmenopausal women. This provides a foundation for implementing risk-based screening recommendations in Asian women.

Analysis of Delirium Risk Assessment Tools for Prediction of Postoperative Delirium Following Lumbar Spinal Fusion.

Retrospective cohort study. Evaluate the utility of Delirium Risk Assessment Score (DRAS), Delirium Risk Assessment Tool (DRAT), and Delirium Elderly At-Risk (DEAR) in patients undergoing posterior lumbar interbody fusions. Surgical interventions can place patients at risk for postoperative delirium (POD), an acute and often severe cognitive impairment associated with poor outcomes. However, common risk assessment tools have not been validated in patients undergoing spine surgery. Adults who underwent posterior lumbar fusion were queried using PearlDiver. Baseline demographics, comorbidities, and delirium occurrence within 7 days of surgery were extracted. Delirium risk scores were calculated using DRAS (15 points total; threshold 5 points), DRAT (8 points total; threshold 3 points), and DEAR (5 points total; threshold 2 points) scales. Receiver operating characteristic (ROC) curves were generated, and optimal risk scores maximizing Youden's Index were established for each measure. Among 37,119 patients, 70 patients (0.2%) developed POD. The mean age was 60.1y, 56.6% were female, and mean Charlson Comorbidity Index (CCI) was 2.1. POD patients had lower mean age and percent female sex, but higher mean CCI and percent medical comorbidities (all P<0.05). ROC curve analyses revealed that a DRAS score of 5 (Sensitivity=62.9%, Specificity=63.9%), DRAT score of 3 (Sensitivity=31.4%, Specificity=81.0%), and DEAR score of 2 (Sensitivity=40.0%, Specificity=82.9%) maximized the Youden's Index value. Patients above these thresholds were 6.0, 2.0, and 3.2 times more likely to develop POD after posterior lumbar fusion, respectively. Delirium risk assessments tools were found to be useful in stratifying patients at high risk of POD following posterior lumbar fusion. Specifically, patients above the pre-defined thresholds were 2 to 6 times more likely to develop delirium postoperatively. Careful stratification of patients' risk of delirium using highly sensitive and specific tools like DRAS may guide preoperative surgical planning and postoperative management plans. IV.

A comparative analysis of risk stratification tools in systemic sclerosis-associated pulmonary arterial hypertension: a EUSTAR analysis.

The 2022 European Society of Cardiology and European Respiratory Society (ESC/ERS) Guidelines for pulmonary arterial hypertension (PAH) recommend risk stratification to optimize management. However, the performance of generic PAH risk stratification tools in patients with systemic sclerosis (SSc)-associated PAH remains unclear. Our objective was to identify the most accurate approach for risk stratification at SSc-PAH diagnosis. In this multicentre, international cohort study from the European Scleroderma Trials and Research (EUSTAR) group database, we screened eleven risk stratification tools upon SSc-PAH diagnosis. We compared the performance of the three top-ranked tools to predict mortality with the ESC/ERS three-strata model, the currently recommended tool for baseline risk assessment. We also assessed the impact of incorporating SSc-specific characteristics into the tools. Kaplan-Meier analyses and Cox regression with area under the ROC curve (AUC) were conducted. The ESC/ERS three-strata model had a lower ability to predict mortality than the ESC/ERS four-strata model, "SPAHR updated", and "REVEAL Lite 2". The ESC/ERS four-strata model divided "intermediate-risk" patients into two groups with significantly different long-term survival rates and is the easiest applicable tool. Incorporating SSc-specific characteristics did not significantly improve the predictive ability of any model, but a low DLCO was an independent predictor of mortality. Considering its ability to predict mortality, risk segregation capabilities, and clinical applicability, this study provides a rationale for using the simplified ESC/ERS four-strata model at SSc-PAH diagnosis as an alternative to the comprehensive ESC/ERS three-strata model. We propose considering DLCO as an individual prognostic marker in SSc-PAH.

Development and external validation of a dynamic nomogram for predicting the risk of functional outcome after 90 days in patients with acute intracerebral hemorrhage

Background and purposeIntracerebral hemorrhage remains a significant cause of death and disability worldwide, highlighting the urgent need for accurate prognostic assessments to optimize patient management. This study aimed to develop a practical nomogram for risk prediction of poor prognosis after 90 days in patients with intracerebral hemorrhage.MethodsA retrospective study was conducted on 638 patients with intracerebral hemorrhage in the Second Hospital of Fujian Medical University, China, who were divided into a training set (n = 446) and a test set (n = 192) by random splitting. Then the data on demographics, clinical symptoms, imaging characteristics, and laboratory findings were collected. In this study, adverse outcomes were defined as a Modified Rankin Scale (mRS) score of 3–6 at 90 days post-ICH onset, as assessed during follow-up. Later, least absolute shrinkage and selection operator (LASSO) regression and multifactorial logistic regression were used to screen the variables and construct a nomogram. Next, the evaluation was performed using the Receiver Operating Characteristic (ROC) curve, calibration curve, and decision curve analysis. Finally, the external validation was completed using the data of 496 patients with intracerebral hemorrhage from the Jinjiang Hospital of Traditional Chinese Medicine.ResultsIn the training and test sets of intracerebral hemorrhage, the incidence of poor prognosis was 60.53 and 61.46%, respectively. Through variable screening, this study identified age, Glasgow Coma Scale (GCS), blood glucose, uric acid, hemoglobin, and hematoma location as independent predictors of poor prognosis in intracerebral hemorrhage. The developed dynamic nomogram was easy to use and demonstrated strong predictive performance (training set AUC: 0.87; test set AUC: 0.839; external validation set AUC: 0.774), excellent calibration, and clinical applicability.ConclusionThe dynamic nomogram we developed using five independent risk factors serves as a practical tool for real-time risk assessment and can help facilitate early intervention and personalized patient management, thereby improving clinical outcomes in high-risk patients.

Biases in the performance of FRAX without BMD in predicting fracture risk in a multiethnic population with diabetes: the Diabetes & Aging Study.

Fracture risk calculators, such as the Fracture Risk Assessment Tool (FRAX), calculate the risk of major osteoporotic (MOF) and hip fracture, but do not account for the excess risk of fracture in people with diabetes. We examined the predictive performance of FRAX without BMD in ethnically diverse, older patients with diabetes. Patients included were between ages 65-89 from the Kaiser Permanente Northern California Diabetes Registry and not already taking osteoporosis medications. Race and ethnicity were self-identified. We calculated FRAX without BMD based on baseline characteristics and assessed how well FRAX predicted MOF and hip fracture over follow-up. Predictive performance was based on measures of discrimination (area under the receiver operator curve, AUC) and calibration (observed-to-predicted ratio, O/P). We identified 96 914 patients (47.0% female), of whom 5383 (5.6%) and 1767 (1.8%) had MOF and hip fracture, respectively, over a mean follow-up of 4.3years. The AUC for MOF and hip fracture were 0.72 and 0.77, respectively. FRAX mildly underestimated MOF and hip fracture rates (O/P 1.2 for both) overall. Discrimination was similar by race and ethnicity and diabetes duration but was worse in those over age 75 (AUC < 0.7). In some groups, there were substantial calibration errors, such as Hispanic women (O/P: 1.8 and 1.5), Black men (O/P: 1.5 and 1.8), those with duration of diabetes ≥20years (O/P: 1.6 and 1.5), and those over the age of 80 (O/P: 1.4 and 1.2) for MOF and hip fracture, respectively. While the discriminatory performance of FRAX without BMD was good overall in patients with diabetes, it underestimated risk in Hispanic women, Black men, those with long duration of diabetes, and in the oldest patients with diabetes. These algorithmic biases suggest that diabetes-specific tools may be needed to stratify fracture risk in patients with diabetes.

P-2299. Febrile neutropenia in cancer: a retrospective analysis in a dominican tertiary hospital

Abstract Background Febrile neutropenia (FN) poses a significant threat to cancer patients, often leading to severe infections. This study aims to characterize the epidemiology and sources of infection among hospitalized cancer patients experiencing FN at Hospital General de la Plaza De la Salud. By elucidating these factors, we strive to enhance patient care and outcome within our context. Source: Statistical registry of the authors. Methods We conducted a retrospective study at a 289-beds in a tertiary care center in the Dominican Republic from March 2021 to March 2024, assessing 258 hospitalized patients with severe neutropenia. 42 cancer patients met the inclusion criteria (absolute neutrophil count less than 500 cells/µL and fever greater than 38.3°C) and data was collected reviewing the eCharts. Source: Statistical registry of the authors. Results 24 patients were male (57.1%), with a median age of 44 (IQR=27). 34 (80.9%) patients had hematologic malignancy, with acute myeloid leukemia being the most common (n=19; 45.2%), followed by non-Hodgkin lymphoma (n=7; 16.7%). Cause of fever was identified for 28 patients, which included bacteremia (13/28; 30.9%), respiratory tract infection (7/28; 16.7%), skin and soft tissue infection (5/28; 11.9%) and others (3/28; 10.7%). An etiological agent was documented in 20 cases, of which the most common was bacterial (15/20;75%); Pseudomonas aeruginosa (6/15; 40%), Escherichia coli (5/15; 33.3%), and Klebsiella pneumoniae (2/15; 20%), 4/20 (20%) were viral; 3/4 due to SARS-COV2 and 1/4 due to dengue, and 1/20 (5%) had Candida albicans. 14 (33.3%) patients had fever of unknown origin. 24/42 patients (68%) had a MASCC score &amp;lt; 21. Mortality rate was 30.9% (n=13), of which 12 (92%) had a MASCC score &amp;lt; 21 (p=0.02). p=0.33. Source: Statistical registry of the authors. Conclusion This study emphasizes how infectious complications greatly impact cancer patients experiencing febrile neutropenia, especially those with hematologic malignancies. Bacterial infections, notably Pseudomonas aeruginosa and Escherichia coli, are prevalent, highlighting the urgency of precise diagnostic methods for tailored antimicrobial treatment. The link between low MASCC scores and higher mortality rates focuses the importance of this risk assessment tool in guiding clinical decisions. These findings stress the necessity for improved multidisciplinary strategies to enhance the outlook for this vulnerable patient group. MASCC risk vs outcome p=0.02 Source: Statistical registry of the authors. Disclosures All Authors: No reported disclosures

P-387. Exploring the Association Between Penicillin Allergy Labels (PAL), β-Lactam Utilization and Surgical Site Infections

Abstract Background Several studies have reported higher surgical site infection (SSI) risk in patients with a penicillin allergy label (PAL). We explore the association between PAL, receipt of β Lactam Antibiotics and SSI in patients undergoing hip and knee arthroplasty.Figure 1: SSI per 100 Hip and Knee Arthroplasty Patients with and without a Penicillin Allergy Label (PAL) Stratified by Receipt of Preoperative β-Lactam Antibiotic Methods We performed a retrospective cohort study of hip (HPRO) and knee arthroplasty (KPRO) patients who underwent procedures in a large health system between August 1, 2022 and December 31, 2023 after implementation of a penicillin allergy risk assessment tool in the electronic health record. We compared demographic, clinical, procedural and SSI outcome data in patients with and without a PAL. Univariate and multivariable logistic regression were performed to assess for factors associated with risk of SSI.Table 1:Demographic and Clinical Characteristics of Hip and Knee Arthroplasty Patients with and without a Penicillin Allergy Label Results Of 9,802 patients who underwent 11,019 HPRO and KPRO procedures, 1851 (19%) reported a penicillin allergy and 102 (1.0%) had one or more SSI. PAL was more common in patients who were older, white, female, lived in areas with socioeconomic status Social Vulnerability Index (SVI) &amp;lt; 75th percentile and who had a higher American Society of Anesthesiologists (ASA) risk score. PAL patients were less likely to receive nasal iodine in the preoperative area. (Table 1) 77% of PAL patients received a preoperative β-lactam for prophylaxis with no reported adverse events. Patients with a PAL were 1.6 times more likely to have a SSI than those without a PAL. The trend toward higher risk of SSI in patients with PAL persisted regardless of whether the patient received preoperative β-lactam antibiotics (Figure 1). In the univariate analysis, day of surgery nasal iodine in the preoperative area and undergoing an outpatient procedure were protective whereas having a HPRO, revision, longer procedure, and higher ASA score were associated with higher odds of SSI. After adjustment for patient age, procedure type, revision, procedure duration, ASA score, outpatient procedure, receipt of a preoperative β-lactam antibiotic and receipt of nasal iodine in the preoperative area, PAL remained associated with 1.6 times higher odds of SSI (p=0.05).Table 2:Comparative Analysis of Unadjusted and Adjusted Odds Ratios for Hip and Knee Arthroplasty Surgical Site Infection Conclusion The increased risk of SSI in patients with PAL may not be fully explained by the lack of preoperative β-lactam antibiotic. More study is needed to assess further. Disclosures All Authors: No reported disclosures

Addressing healthcare disparities and improving osteoporosis management in rural communities: a cluster randomized control trial.

Rural communities face limited healthcare access, financial constraints, and transportation barriers leading to health disparities. This study examined interventions that reduced health disparities in increasing the outpatient attendance and treatment rate of anti-osteoporosis medication (AOM), while identifying factors contributing to therapy refusal in rural communities. A total of 567 patients were randomized at the community level into three groups: multicomponent integrated care (MIC), osteoporosis care only (OC), and usual care (UC). Fracture Risk Assessment Tool and dual-energy X-ray absorptiometry scans were used to evaluate the osteoporosis and osteoporotic fracture risk. High- and moderate-risk patients were advised to pursue further hospital-based assessments and treatment. Both the MIC and OC groups received five interventions to address rural barriers, including specialist access, disease education, overcoming transportation barriers, peer support, and dedicated case managers. However, UC excluded transportation assistance, peer support, and case management. Outcomes measured included outpatient attendance, AOM treatment rates, and factors affecting hospital assessment refusal, analyzed via multivariable logistic modeling. In the MIC group, 73.3% of patients attended the outpatient clinic and 58.6% received AOM. In the OC group, 81% patients attended and 69.3% received AOM. Conversely, in the UC group, only 4.1% attended and received AOM. Significant differences in attendance and AOM rates were found between the MIC and UC groups and between the OC and UC groups (p < .001 for both). Common barriers included beliefs that treatment was unnecessary and lack of hospital access. Risk factors hindering outpatient attendance include male sex, low education, low budget, multiple disabilities, and osteopenia diagnosis. Addressing transportation barriers and implementing dedicated case management are crucial for improving healthcare access among rural patients. ClinicalTrials.gov NCT05104034.

International validation of a pre-transplant risk assessment tool for graft survival in pediatric kidney transplant recipients

Abstract Background A pre-transplant prediction model using commonly available factors is valuable for optimizing donor selection, communication, and counselling for pediatric kidney transplant (PKT) recipients. This study aims to externally validate a Dutch PKT prediction model and assess its international applicability. Methods Data from the Dutch-, CERTAIN-, and CRISTAL registries, covering PKT from 2005 to 2021, were used. The Dutch prediction model was externally validated in a German and French cohort and then adapted to these specific countries. An international prediction model was also developed using all available data. Models were based on 80% derivation cohorts and internally validated using areas under the receiver operating characteristics curve (ROC-AUC) and calibration plots. Results Of 3266 transplantations, 2475 (273 Dutch, 356 German and 1622 French, 224 other) were used for analysis. Cohorts differed significantly in baseline characteristics and outcomes. Internal validation of the Dutch model showed ROC-AUCs of 0.77 and 0.75 at 10 and 15 years. External validation in German and French cohorts yielded 10-year ROC-AUCs of 0.63 and 0.60, respectively. Internal validation of the international prediction model showed AUCs of 0.61 and 0.60 at 10 and 15 years with poor calibration, indicating inferior performance. The adapted national models showed better internal validation performance, with 10-year ROC-AUCs of 0.77, 0.76, and 0.73 in Dutch, French, and German cohorts, respectively. Conclusions The Dutch PKT prediction tool requires country-specific adaptations for use in other countries, given the diversity of clinical practice across Europe. A country-specific model is preferable to an international model in the current landscape.

A multi-model machine learning framework for breast cancer risk stratification using clinical and imaging data

Purpose This study presents a comprehensive machine learning framework for assessing breast cancer malignancy by integrating clinical features with imaging features derived from deep learning. Methods The dataset included 1668 patients with documented breast lesions, incorporating clinical data (e.g., age, BI-RADS category, lesion size, margins, and calcifications) alongside mammographic images processed using four CNN architectures: EfficientNet, ResNet, DenseNet, and InceptionNet. Three predictive configurations were developed: an imaging-only model, a hybrid model combining imaging and clinical data, and a stacking-based ensemble model that aggregates both data types to enhance predictive accuracy. Twelve feature selection techniques, including ReliefF and Fisher Score, were applied to identify key predictive features. Model performance was evaluated using accuracy and AUC, with 5-fold cross-valida tion and hyperparameter tuning to ensure robustness. Results The imaging-only models demonstrated strong predictive performance, with EfficientNet achieving an AUC of 0.76. The hybrid model combining imaging and clinical data reached the highest accuracy of 83% and an AUC of 0.87, underscoring the benefits of data integration. The stacking-based ensemble model further optimized accuracy, reaching a peak AUC of 0.94, demonstrating its potential as a reliable tool for malignancy risk assessment. Conclusion This study highlights the importance of integrating clinical and deep imaging features for breast cancer risk stratification, with the stacking-based model.