Clinical Decision Aid Research Articles

Early prediction of unfavorable evolution after a first clinical episode suggestive of multiple sclerosis: the EUMUS score.

Predicting disease progression in patients with the first clinical episode suggestive of multiple sclerosis (MS) is crucial for personalized therapeutic approaches. This study aimed to develop the EUMUS score for accurately estimating the risk of early evidence of disease activity and progression (EDA). Retrospective analysis was conducted on data from 221 patients with a first clinical MS episode collected from four Italian MS centers. Various variables including socio-demographics, clinical features, cerebrospinal fluid analysis, evoked potentials, and brain MRI were considered. A prognostic multivariate regression model was identified to develop the EUMUS score. The optimal cutoff for predicting the transition from no evidence of disease activity (NEDA3) to EDA was determined. The accuracy of the prognostic model and score were tested in a separate UK MS cohort. After 12 months, 61.54% of patients experienced relapses and/or new MRI lesions. Younger age (OR 0.96, CI 0.93-0.99; p = 0.005), MRI infratentorial lesion(s) at baseline (OR 2.21, CI 1.27-3.87; p = 0.005), positive oligoclonal bands (OR 2.89, CI 1.47-5.69; p = 0.002), and abnormal lower limb somatosensory-evoked potentials (OR 2.77, CI 1.41-5.42; p = 0.003) were significantly associated with increased risk of EDA. The EUMUS score demonstrated good specificity (72%) and correctly classified 80% of patients with EDA in the independent UK cohort. The EUMUS score is a simple and useful tool for predicting MS evolution within 12 months of the first clinical episode. It has the potential to guide personalized therapeutic approaches and aid in clinical decision-making.

Development of The Lupus Clinical Trials Enrollment Decision Aid: a pilot study

IntroductionIn this pilot study, we describe the development of a patient-centered Decision Aid (DA) for participation of SLE clinical trials called “The Lupus Clinical Trials Enrollment DA”.MethodsA draft DA was designed by a development working group using a collaborative, iterative process using the International Patient Decision Aid Standards (IPDAS) guidelines. The approved draft DA was then pilot tested and refined using semi structured interview with 10 lupus providers and 12 SLE patients. Descriptive statistics were calculated. Interviews/surveys were conducted until thematic saturation was achieved. Responses on usefulness were accumulated, and mean usefulness scores were calculated. Feedback from the semi-structured interviews were categorized into several themes as outlined in the results section.ResultsThe definition of treatments, side effects of each option, and expected improvement from each option was outlined. 90% of providers and 91.7% of patients reported that the definition of SOC treatment was clear. Additionally, the expected improvement for SOC (90% of providers, 100% of patients), clinical trial drug (70%, 91.6%), and placebo (70%, 100%) were noted to be clear. Side effects of SOC (80%, 100%), clinical trial drug treatment (80%, 100%), placebo (90%, 100%), were also noted to be clear. 100% of providers and patients thought that the figure outlining pros/cons of participating in clinical trials was appropriate. The mean usefulness scores for the DA were 4.45/5 for providers and 4.67/5 for patients.DiscussionThese data demonstrate that both patients and providers confirm that the newly developed The Lupus Clinical Trials Enrollment DA is useful and easy to use. Qualitative feedback from providers demonstrated concern that aspects of the DA, such as expected improvement and side effects might be unclear to patients; however, patients did not express the same concern in either the quantitative or qualitative feedback.

Evaluating Hymenoptera Venom Allergy Severity: A Data-Centric Comparison of Grading Instruments

Introduction: While a consensus seems to have been reached with regard to the definition of anaphylaxis, there is no universal instrument for scoring allergic reaction severity despite more than 30 having been proposed by the time of writing. This severely hampers comparison of data between studies. While scales have been compared with regard to their utility in grading food-related reactions, no such comparisons have been made for Hymenoptera venom-associated reactions. Methods: The study conducted a retrospective analysis to compare the severity of Hymenoptera venom allergy reactions in 104 participants with suspected Hymenoptera venom allergy. The study applied six grading instruments to each reaction, also evaluating them against the NIAID/FAAN anaphylaxis criteria. Sensitivity, specificity, and receiver operating characteristic area under the curve (AUC) for identifying anaphylaxis were calculated. Severity scales were simplified into “mild,” “moderate,” and “severe” categories. The most common severity grade across the five scales was determined using a custom function to establish a consensus severity grade. Results: The most common culprit insects were honeybees (49.0%). Among the 88 participants with generalized reactions, the highest proportion had involvement of four organ systems. The scales showed high specificity for detecting anaphylaxis, especially when using higher grades of the Mueller, WAO, and Dribin scales. The diagnostic yields (AUC) varied, with the WAO scale having the highest AUC (0.94) for grades 3, 4, and 5. Spearman correlation analysis showed the strongest correlations seen between the Brown and Dribin, Ring and Messmer and Dribin, and Ring and Messmer and Reisman scales. The lowest correlations were observed with the Mueller scale when paired with the WAO, Reisman, and Dribin scales. An inter-rater reliability analysis showed substantial agreement between scales with the same number of grading levels. The agreement was highest for the Brown and Dribin scales, indicating a strong consistency in reaction severity classification across different instruments. Conclusion: While all instruments were effective in stratifying reactions, they showed limitations in differentiating milder phenotypes. The Brown and Dribin scales stood out for their high agreement with the consensus score and sensitivity in identifying anaphylaxis. Our findings suggest that adopting either of these scales could significantly unify the reporting of allergic reactions. We believe the format of an instrument should be tailored to its intended purpose, with clinical decision aids being simpler and research tools being more detailed.

A Literature Review for Detection and Projection of Cardiovascular Disease Using Machine Learning

The heart is a vital organ that is indispensable in ensuring the general health and welfare of individuals. Cardiovascular diseases (CVD) are the major health concern worldwide and a leading cause of death, leaving behind diabetes and cancer. To deal with the problem, it is essential for early detection and prediction of CVDs, which can significantly reduce morbidity and mortality rates. Computer-aided techniques facilitate physicians in the diagnosis of many heart disorders, such as valve dysfunction, heart failure, etc. Living in an "information age," every day million bytes of data are generated, and we can turn these data into knowledge for clinical investigation using the technique of data mining. Machine learning algorithms have shown promising results in predicting heart disease based on different risk parameter. In this study, for the purpose of predicting CVDs, our aim is to appraise and examine the outputs generated by machine learning algorithms including support vector machines, artificial neural network, logistic regression, random forest and decision trees.This literature survey highlights the correctness of different machine learning algorithms in forecasting heart problem and can be used as a basis for building a Clinical decision-making aid to detect and prevent heart disease at an early stage.

Antimicrobial stewardship markers and healthcare-associated pneumonia threshold criteria in UK hospitals: analysis of the MicroGuideTm application.

To address antimicrobial resistance, antimicrobial stewardship (AMS) principles must be implemented and adhered to. Clinical decision aids such as the MicroGuideTM app are an important part of these efforts. We sought to evaluate the consistency of core AMS information and the diversity of classification thresholds for healthcare-associated pneumonia (HAP) in the MicroGuide app. Guidelines in the MicroGuide app were extracted and analysed for content related to AMS and HAP. Guidelines were characterized according to HAP naming classification; community-acquired pneumonia (CAP) classifications were analysed to serve as a comparator group. In total, 115 trusts (119 hospitals) were included. Nearly all hospitals had developed MicroGuide sections on AMS (n = 112/119, 94%) and sepsis management (n = 117/119, 98%). Other AMS sections were outpatient parenteral antimicrobial therapy (47%), antifungal stewardship (70%), critical care (23%) and IV to oral switch therapy (83%). Only 9% of hospitals included guidance on the maximum six key AMS sections identified. HAP definitions varied widely across hospitals with some classifying by time to onset and some classifying by severity or complexity. The largest proportion of HAP guidelines based classification on severity/complexity (n = 69/119, 58%). By contrast, definitions in CAP guidelines were uniform. The high heterogeneity in HAP classification identified suggests inconsistency of practice in identifying thresholds for HAP in the UK. This complicates HAP management and AMS practices. To address HAP in alignment with AMS principles, a comprehensive strategy that prioritizes uniform clinical definitions and thresholds should be developed.

Risk factors and predictive model for prenatal depression: A large retrospective study in China

BackgroundPrenatal depression, associated with adverse effects on mothers and fetuses, has received little attention. We conducted a large-sample study to investigate the risk factors of, and develop a predictive model for, prenatal depression in the Chinese population. MethodsThis study enrolled 14,329 pregnant women who delivered at the West China Second University Hospital, Sichuan University from January 2017 to December 2020. Participants were divided into a training or validation cohort. Multiple variables were collected and selected using univariate logistic regression and least absolute shrinkage and selection operator penalty regression. After multivariate logistic analysis, a predictive model was developed and validated internally and externally. ResultsNine variables (employment, planned pregnancy, pregnancy number, conception methods, gestational diabetes mellitus, twin pregnancy, placenta previa, umbilical cord encirclement, and educational attainment) were identified as independent risk factors for prenatal depression. Receiver operating characteristic curves in both the training and validation cohorts showed excellent discrimination of the predictive model (the area under the curve: 0.746 and 0.732, respectively). LimitationsThe results of this retrospective study may be affected by confounding and information bias. Some important variables were excluded, such as family history of mental disorders. The study was conducted in China; its results may not be generalizable to other regions. ConclusionOur study identified nine significant risk factors for prenatal depression and constructed an accurate predictive model. This model could be applied as a clinical decision aid for individualized risk estimates and prevention of prenatal depression.

Recommendations to address respondent burden associated with patient-reported outcome assessment.

Patient-reported outcomes (PROs) are increasingly used in healthcare research to provide evidence of the benefits and risks of interventions from the patient perspective and to inform regulatory decisions and health policy. The use of PROs in clinical practice can facilitate symptom monitoring, tailor care to individual needs, aid clinical decision-making and inform value-based healthcare initiatives. Despite their benefits, there are concerns that the potential burden on respondents may reduce their willingness to complete PROs, with potential impact on the completeness and quality of the data for decision-making. We therefore conducted an initial literature review to generate a list of candidate recommendations aimed at reducing respondent burden. This was followed by a two-stage Delphi survey by an international multi-stakeholder group. A consensus meeting was held to finalize the recommendations. The final consensus statement includes 19 recommendations to address PRO respondent burden in healthcare research and clinical practice. If implemented, these recommendations may reduce PRO respondent burden.

Designing an Implementable Clinical Prediction Model for Near-Term Mortality and Long-Term Survival in Patients on Maintenance Hemodialysis

The life expectancy of patients treated with maintenance hemodialysis (MHD) is heterogeneous. Knowledge of life-expectancy may focus care decisions on near-term versus long-term goals. The current tools are limited and focus on near-term mortality. Here, we develop and assess potential utility for predicting near-term mortality and long-term survival on MHD. Predictive modeling study. 42,351 patients contributing 997,381 patient months over 11 years, abstracted from the electronic health record (EHR) system of midsize, nonprofit dialysis providers. Demographics, laboratory results, vital signs, and service utilization data available within dialysis EHR. For each patient month, we ascertained death within the next 6 months (ie, near-term mortality) and survival over more than 5 years during receipt of MHD or after kidney transplantation (ie, long-term survival). We used least absolute shrinkage and selection operator logistic regression and gradient-boosting machines to predict each outcome. We compared these to time-to-event models spanning both time horizons. We explored the performance of decision rules at different cut points. All models achieved an area under the receiver operator characteristic curve of≥0.80 and optimal calibration metrics in the test set. The long-term survival models had significantly better performance than the near-term mortality models. The time-to-event models performed similarly to binary models. Applying different cut points spanning from the 1st to 90th percentile of the predictions, a positive predictive value (PPV) of 54% could be achieved for near-term mortality, but with poor sensitivity of 6%. A PPV of 71% could be achieved for long-term survival with a sensitivity of 67%. The retrospective models would need to be prospectively validated before they could be appropriately used as clinical decision aids. A model built with readily available clinical variables to support easy implementation can predict clinically important life expectancy thresholds and shows promise as a clinical decision support tool for patients on MHD. Predicting long-term survival has better decision rule performance than predicting near-term mortality. Clinical prediction models (CPMs) are not widely used for patients undergoing maintenance hemodialysis (MHD). Although a variety of CPMs have been reported in the literature, many of these were not well-designed to be easily implementable. We consider the performance of an implementable CPM for both near-term mortality and long-term survival for patients undergoing MHD. Both near-term and long-term models have similar predictive performance, but the long-term models have greater clinical utility. We further consider how the differential performance of predicting over different time horizons may be used to impact clinical decision making. Although predictive modeling is not regularly used for MHD patients, such tools may help promote individualized care planning and foster shared decision making.

Abstract WP203: Prediction of Outcome After Endovascular Therapy for Acute Ischemic Stroke: A Machine Learning Approach From Japan Stroke Data Bank

Introduction: Previous studies have demonstrated the predictive capabilities of machine learning (ML) models using clinical, imaging, and angiographic characteristics in determining the clinical outcome of patients undergoing endovascular treatment for acute ischemic stroke. We aimed to analyze a large-scale stroke registry and develop an ML model that solely utilizes clinical characteristics, without the need for imaging or angiographic data, to predict the outcome of patients treated with endovascular therapy for acute ischemic stroke. Methods: We conducted our analysis using data from the Japan Stroke Data Bank, a nationwide acute stroke registry containing individual patient data. A total of 4353 patients who underwent endovascular treatment for acute ischemic stroke between January 2016 and December 2020 were included in the study. Patient data from 2016 to 2019 were used as the derivation cohort, while data from 2020 were used as the validation cohort. An ML models were trained on 70% of the patients randomly selected from the derivation cohort and tested on the remaining 30%. The input variables for the ML models consisted of 138 clinical characteristics including neurological examination findings. Imaging, angiographic, and laboratory data were not utilized in the ML process. The primary outcome measure was good outcome at discharge, defined as a modified Rankin Scale score of 0-2. Results: Of the 4353 patients included in the study, 1520 patients (34.9%) achieved good outcome. The area under the curve for predicting good outcome was 0.76 for the test set in the derivation cohort and 0.78 for the validation cohort. The SHAP summary plot revealed that the top five major predictive contributors for good outcomes were pre-stroke modified Rankin Scale score, level of consciousness, age, stroke classification, and NIHSS score. Conclusions: Our findings demonstrate that an ML model utilizing solely clinical characteristics, without the inclusion of imaging or angiographic data, can accurately predict good outcomes at discharge for acute stroke patients treated with endovascular therapy. This approach has the potential to provide an aid in clinical decision-making before imaging examination.

Emerging Research Interest in Medical and Dental Undergraduates

I am writing this letter to highlight the current encouraging and commendable rise in the involvement of undergraduate medical and dental students in research across Pakistan. A positive surge was detected among the medical and dental institutes reflecting increasing interest and contribution in research activities. A study in Saudi Arabia reported the relationship between rising research trends and its contributing factors, henceforth, stating that admission into residency (56.8%), interest in research (41.8%), and aid in clinical decision-making (31%) were the top three factors for research interest 1.

Use of the Capability, Opportunity and Motivation Behaviour model (COM-B) to Understand Interventions to Support Physical Activity Behaviour in People with Stroke: An Overview of Reviews.

Physical activity in people with stroke remains low despite considerable research. This overview aimed to provide high-level synthesis and aid clinical decision-making. The Capability, Opportunity, Motivation-Behaviour (COM-B) model was used to classify interventions to understand which components improve physical activity behaviour in people with stroke. CINAHL, Cochrane Database, MEDLINE, PEDro, PsychINFO, SPORTDiscus. A systematic search was conducted (November 2023) to identify reviews of interventions to improve physical activity in people with stroke. Results were screened and assessed for eligibility. Participant characteristics, intervention classification using COM-B, and effect of intervention were extracted. Quality was assessed using AMSTAR2, and Corrected Cover Analysis for study overlap. Narrative synthesis was used to understand components of interventions to improve physical activity behaviour. 1801 references were screened and 29 full-text references assessed for eligibility. Twenty reviews were included. Quality ranged from critically low (n = 3) to high (n = 10). Study overlap calculated using corrected cover area indicated slight overlap (0.028) and minimal reporting bias.The majority of participants were mobile with mild stroke and community dwelling. Twenty-three interventions were classified using COM-B. Three of twelve interventions classified to one aspect of the COM-B were effective. Fourteen of sixteen effective interventions combined at least two COM-B elements, ten of these combined capability and motivation. Interventions including at least two elements of the COM-B are most likely to improve physical activity in mobile stroke survivors. Further research is needed to understand physical activity behaviour in those with moderate to severe stroke.

Impact of Remote Patient Monitoring Platform on Patients With Moderate to Severe Persistent Asthma: Observational Study.

Asthma is one of the most common respiratory diseases, with an ever-growing health care burden. Remote patient monitoring (RPM) has gained increasing importance in the respiratory care area with the outbreak of the COVID-19 pandemic. In this pilot study, we introduced a novel platform that remotely monitors patients with chronic respiratory illnesses using Centers for Disease Control and Prevention guidelines to reduce hospitalizations and emergency department visits. This study aimed to understand patient and physician engagement with a new virtual care solution (KevaTalk app and Keva365 platform) and the value, for both patients and providers, of using an RPM tool. We assessed real-world use of the platform from both physician and patient perspectives and the impact of devices on engagement and monitoring. Participants with a history of moderate to severe persistent asthma, seen by a pulmonologist at a hospital, were included in this study. The inclusion criteria involved being aged ≥18 years and having access to an Android or iOS mobile device with internet. We provided patient questionnaires to assess the app's usefulness and evaluate its features. We monitored remote spirometry and oximetry data, app check-ins, alerts, and escalations during this study's time window. Data were reviewed daily and predetermined criteria were set to escalate for physician review based on the patient's symptoms and objective data. Overall, 25 patients were included in this pilot. The mean age was 57 (SD 10.7) years and a majority (n=23, 92%) were female. A baseline questionnaire, which was used to rate the app, indicated that the ease of check-in and ease of modification to the patient's asthma plan were the 2 highest rated features. In total, 2066 check-ins (1550 green, 506 yellow, and 10 red check-ins) and 1155 spirometry sessions were recorded during this 3-month period. Further, 64% (14/22) and 91% (20/22) of patients were found to have peak flows in their red and yellow zones at least once, respectively. During the course of this study, 484 alerts were recorded and evaluated by the team, of which 37.2% (n=180) required an escalation to the physician; this included a transfer to a medical facility, change in respiratory medication, or further education. In this pilot study, we demonstrated the feasibility of implementing a novel RPM platform in patients with asthma. Our platform showed high patient engagement and satisfaction and provided physicians with real-time subjective data to evaluate patients remotely that aids in clinical decision-making. The escalations prevented patients from having an exacerbation or flare up, which led to the prevention of an emergency department visit. Continuous monitoring of chronic disease has benefits over episodic monitoring. It allows for improved quality of life, better outcomes, and huge health care savings.

A dedicated risk prediction model of 1-year mortality following endovascular aortic aneurysm repair involving the renal-mesenteric arteries

ObjectiveTreatment goals of prophylactic endovascular aortic repair of complex aneurysms involving the renal-mesenteric arteries (complex endovascular aortic repair [cEVAR]) include achieving both technical success and long-term survival benefit. Mortality within the first year after cEVAR likely indicates treatment failure owing to associated costs and procedural complexity. Notably, no validated clinical decision aid tools exist that reliably predict mortality after cEVAR. The purpose of this study was to derive and validate a preoperative prediction model of 1-year mortality after elective cEVAR. MethodsAll elective cEVARs including fenestrated, branched, and/or chimney procedures for aortic disease extent confined proximally to Ishimaru landing zones 6 to 9 in the Society for Vascular Surgery Vascular Quality Initiative were identified (January 2012 to August 2023). Patients (n = 4053) were randomly divided into training (n = 3039) and validation (n = 1014) datasets. A logistic regression model for 1-year mortality was created and internally validated by bootstrapping the AUC and calibration intercept and slope, and by using the model to predict 1-year mortality in the validation dataset. Independent predictors were assigned an integer score, based on model beta-coefficients, to generate a simplified scoring system to categorize patient risk. ResultsThe overall crude 1-year mortality rate after elective cEVAR was 11.3% (n = 456/4053). Independent preoperative predictors of 1-year mortality included chronic obstructive pulmonary disease, chronic renal insufficiency (creatinine >1.8 mg/dL or dialysis dependence), hemoglobin <12 g/dL, decreasing body mass index, congestive heart failure, increasing age, American Society of Anesthesiologists class ≥IV, current tobacco use, history of peripheral vascular intervention, and increasing extent of aortic disease. The 1-year mortality rate varied from 4% among the 23% of patients classified as low risk to 23% for the 24% classified as high risk. Performance of the model in validation was comparable with performance in the training data. The internally validated scoring system classified patients roughly into quartiles of risk (low, low/medium, medium/high and high), with 52% of patients categorized as medium/high to high risk, which had corresponding 1-year mortality rates of 11% and 23%, respectively. Aneurysm diameter was below Society for Vascular Surgery recommended treatment thresholds (<5.0 cm in females, <5.5 cm in males) in 17% of patients (n = 679/3961), 41% of whom were categorized as medium/high or high risk. This subgroup had significantly increased in-hospital complication rates (18% vs 12%; P = .02) and 1-year mortality (13% vs 5%; P < .0001) compared with patients in the low- or low/medium-risk groups with guideline-compliant aneurysm diameters (≥5.0 cm in females, ≥5.5 cm in males). ConclusionsThis validated preoperative prediction model for 1-year mortality after cEVAR incorporates physiological, functional, and anatomical variables. This novel and simplified scoring system can effectively discriminate mortality risk and, when applied prospectively, may facilitate improved preoperative decision-making, complex aneurysm care delivery, and resource allocation.

Real-world reported adverse events related to systemic immunomodulating therapy in patients with atopic dermatitis: Results from the TREAT NL (TREatment of ATopic eczema, the Netherlands) registry.

Evidence on the (long-term) safety of systemic immunomodulating therapies in atopic dermatitis (AD) generated by real-world data is sparse. To describe real-world reported adverse drug reactions (AEs) related to systemic immunomodulating therapy in patients with AD and to compare the incidence rates of AEs with the Summaries of Product Characteristics (SmPCs). We conducted an observational prospective multi-centre cohort study, using the TREAT NL registry. All severe AEs, AEs of special interest and serious AEs in adult and paediatric patients on systemic immunomodulating treatment (ciclosporin, methotrexate, azathioprine, mycophenolic acid, dupilumab, tralokinumab, baricitinib and upadacitinib) were assessed. Incidences rates of all (potentially) drug-related AEs were standardized in patient years and compared to the cumulative incidences in the associated SmPCs. We collected 422 patient years of safety data from 266 patients, of whom 129 (48.5%) reported a total of 224 (potentially) drug-related AEs. Compared to dupilumab's SmPC, higher incidence rates were found for four AEs (reported ≥5 times): eosinophilia, blepharitis, dry eyes and head and neck erythema (i.e. dupilumab facial redness). A higher incidence rate of fatigue was found in patients on oral methotrexate in our cohort compared to the SmPC. Two new drug-related AEs (reported ≥5 times) were found in patients on dupilumab, including non-infectious conjunctivitis and meibomian gland dysfunction. Real-world reported AEs captured in AD patient registries can add information on the estimated incidence of AEs and benefit clinical decision aids. Future studies using data derived from the TREAT NL registry combined with data from other registries within the TREAT Registry Taskforce will provide more information on (rare) AEs associated with immunomodulating therapy in AD patients.

Online interpretable dynamic prediction models for postoperative delirium after cardiac surgery under cardiopulmonary bypass developed based on machine learning algorithms: A retrospective cohort study

ObjectivePostoperative delirium (POD) is strongly associated with poor early and long-term prognosis in cardiac surgery patients with cardiopulmonary bypass (CPB). This study aimed to develop dynamic prediction models for POD after cardiac surgery under CPB using machine learning (ML) algorithms. MethodsFrom July 2021 to June 2022, clinical data were collected from patients undergoing cardiac surgery under CPB at Nanjing First Hospital. A dataset from the same center (October 2022 to November 2022) was also used for temporal external validation. We used ML and deep learning to build models in the training set, optimized parameters in the test set, and finally validated the best model in the validation set. The SHapley Additive exPlanations (SHAP) method was introduced to explain the best models. ResultsOf the 885 patients enrolled, 221 (25.0%) developed POD. 22 (22.0%) of 100 validation cohort patients developed POD. The preoperative and postoperative artificial neural network (ANN) models exhibited optimal performance. The validation results demonstrated satisfactory predictive performance of the ANN model, with area under the receiver operator characteristic curve (AUROC) values of 0.776 and 0.684 for the preoperative and postoperative models, respectively. Based on the ANN algorithm, we constructed dynamic, highly accurate, and interpretable web risk calculators for POD. ConclusionsWe successfully developed online interpretable dynamic ANN models as clinical decision aids to identify patients at high risk of POD before and after cardiac surgery to facilitate early intervention or care.

Evaluating the use of clinical decision aids in an Australian emergency department: A cross-sectional survey.

To identify healthcare professionals' knowledge, self-reported use, and documentation of clinical decision aids (CDAs) in a large ED in Australia, to identify behavioural determinants influencing the use of CDAs, and healthcare professionals preferences for integrating CDAs into the electronic medical record (EMR) system. Healthcare professionals (doctors, nurses and physiotherapists) working in the ED at the Gold Coast Hospital, Queensland were invited to complete an online survey. Quantitative data were analysed using descriptive statistics, and where appropriate, mapped to the theoretical domains framework to identify potential barriers to the use of CDAs. Qualitative data were analysed using content analysis. Seventy-four healthcare professionals (34 medical officers, 31 nurses and nine physiotherapists) completed the survey. Healthcare professionals' knowledge and self-reported use of 21 validated CDAs was low but differed considerably across CDAs. Only 4 out of 21 CDAs were reported to be used 'sometimes' or 'always' by the majority of respondents (Ottawa Ankle Rule for ankle injury, Wells' criteria for pulmonary embolism, Wells' criteria for deep vein thrombosis and PERC rule for pulmonary embolism). Most respondents wanted to increase their use of valid and reliable CDAs and supported the integration of CDAs into the EMR to facilitate their use and support documentation. Potential barriers impacting the use of CDAs represented three theoretical domains of knowledge, social/professional role and identity, and social influences. CDAs are used variably by healthcare professionals and are inconsistently applied in the clinical encounter. Preferences of healthcare professionals need to be considered to allow the successful integration of CDAs into the EMR.

Predictors and prognosis of tracheostomy in relapsing polychondritis.

Airway obstruction can occur in patients with relapsing polychondritis (RP) with laryngeal involvement, occasionally requiring tracheostomy to avoid serious complications. Herein, we assessed the risk factors for tracheostomy and developed a risk prediction model. Clinical characteristics of patients with RP, with and without tracheostomy, were compared using multivariate logistic regression analysis to identify risk factors. A nomogram was developed to predict the population at risk of requiring tracheostomy. In total, 232 patients with RP were reviewed, of whom 146 had laryngeal involvement. Among them, 21 underwent a tracheostomy. Multivariate logistic analysis identified ages ≤ 25 or ≥ 65 (p< 0.001, OR: 24.584, 95% CI: 5.310-113.815), laryngotracheal oedema (p< 0.001, OR: 26.685, 95% CI: 4.208-169.228), and pulmonary infection (p= 0.001, OR: 18.834, 95% CI: 3.172-111.936) as independent risk factors for tracheostomy. A nomogram with a C-index of 0.936 (95% CI: 0.894-0.977) was established based on the multivariate analysis. Internal bootstrap resampling (1000 repetitions) confirmed sufficient discriminatory power with a C-index of 0.926. Decision curve analysis indicated a superior net benefit of the nomogram. Tracheostomy was associated with a significant increase in the in-hospital mortality rate (p= 0.021), but it did not affect the long-term survival rate (p= 0.706). Tracheostomy is associated with an increase in the short-term mortality rate but does not affect the long-term survival rate. The nomogram developed in this study may help identify patients at high risk for tracheostomy and aid in clinical decision-making.

Physician views of artificial intelligence in otolaryngology and rhinology: A mixed methods study.

The study aimed to investigate otolaryngologists' knowledge, trust, acceptance, and concerns with clinical applications of artificial intelligence (AI). This study used mixed methods with survey and semistructured interviews. Survey was e-mailed to American Rhinologic Society members, of which a volunteer sample of 86 members responded. Nineteen otolaryngologists were purposefully recruited and interviewed until thematic saturation was achieved. Seventy-six respondents (10% response rate) completed the majority of the survey: 49% worked in academic settings and 43% completed residency 10 or fewer years ago. Of 19 interviewees, 58% worked in academic settings, and 47% completed residency 10 or fewer years ago. Familiarity: Only 8% of survey respondents reported having AI training in residency, although 72% had familiarity with general AI concepts; 0 interviewees had personal experience with AI in clinical settings. Expected uses: Of the surveyed otolaryngologists, 82% would use an AI-based clinical decision aid and 74% were comfortable with AI proposing treatment recommendations. However, only 44% of participants would trust AI to identify malignancy and 53% to interpret radiographic images. Interviewees trusted AI for simple tasks, such as labeling septal deviation, more than complex ones, such as identifying tumors. Factors influencing AI adoption: 89% of survey participants would use AI if it improved patient satisfaction, 78% would be willing to use AI if experts and studies validated the technologies, and 73% would only use AI if it increased efficiency. Sixty-one percent of survey respondents expected AI incorporation into clinical practice within 5 years. Interviewees emphasized that AI adoption depends on its similarity to their clinical judgment and to expert opinion. Concerns included nuanced or complex cases, poor design or accuracy, and the personal nature of physician-patient relationships. Few physicians have experience with AI technologies but expect rapid adoption in the clinic, highlighting the urgent need for clinical education and research. Otolaryngologists are most receptive to AI "augmenting" physician expertise and administrative capacity, with respect for physician autonomy and maintaining relationships with patients. Level VI, descriptive or qualitative study.

Development and validation of a prognostic nomogram for the renal relapse of lupus nephritis

ObjectivesThis study aims to assess the risk of relapse after complete remission (CR) and partial remission (PR), and to develop a prognostic nomogram predicting the probability in lupus nephritis (LN) patients. MethodsData from patients with LN who had been in remission were collected as a training cohort. The prognostic factors were analyzed using the univariable and multivariable Cox model for the training group. A nomogram was then developed using significant predictors in multivariable analysis. Both discrimination and calibration were assessed by bootstrapping with 100 resamples. ResultsA total of 247 participants were enrolled, including 108 in the relapse group and 139 in the no relapse group. In multivariate Cox analysis, Systemic Lupus Erythematosus Disease Activity Index (SLEDAI), erythrocyte sedimentation rate (ESR), complement 1q (C1q), and antiphospholipid (aPL), anti-Sm antibody were found to be significant for predicting relapse rates. The prognostic nomogram including the aforementioned factors effectively predicted 1- and 3-year probability of flare-free. Moreover, a favorable consistency between the predicted and actual survival probabilities was demonstrated using calibration curves. ConclusionsHigh SLEDAI, ESR, and positive aPL, anti-Sm antibody are potential risk factors for LN flare, while high C1q can reduce its recurrence. The visualized model we established can help predict the relapse risk of LN and aid clinical decision-making for individual patients.

A-122 Fluid Type (STILL) Matters: An Analysis of Quality Control Material Suitability for Measuring Eleven Body Fluid Analytes in the Clinical Laboratory

Abstract Background Measuring the concentration of albumin, amylase, bilirubin, cholesterol, creatinine, glucose, lactate dehydrogenase (LDH), lipase, total protein(TP), triglycerides, and urea nitrogen(UN) in body fluids(BFs) provides important information to aid in clinical decision-making. The range of expected values of these analytes in BFs may not match serum, and the distribution by fluid type is not well described. Quality control (QC) materials are designed to reflect clinically relevant concentrations in blood. Hence, the measuring ranges for assays modified for BF testing vary in suitability. Laboratorians and manufacturers would benefit from understanding the overlap in BF and serum analyte concentrations. The aim of this study was to characterize the range of BF analyte concentrations compared to serum and assess the suitability of QC materials used on these assays. Methods A retrospective analysis was performed for the following analytes in BFs with reportable ranges(including maximum dilution factor): albumin = 0.2–12.0 g/dL(×2), amylase = 3–14 000 U/L(×100), bilirubin = 0.1–72.2 mg/dL(×2), cholesterol = 4–8000 mg/dL(×10), creatinine = 0.1–92.0 mg/dL(×4), glucose = 2–3375 mg/dL(×5), LDH = 10–9000 U/L(×10), lipase = 3–300 000 U/L(×1000), TP = 0.2–36.0 g/dL(×3), triglycerides = 4–8850 mg/dL(×10), and UN = 2–300 mg/dL(×3). Chemistry tests were run on a cobas® 6000 c501 (Roche Diagnostics,Indiana). Patient analyte concentration and fluid type results reported between 9/16/2019 and 1/27/2022 at all Mayo Clinic locations were extracted from an institutional database of electronic medical record data. Serum results were included for comparison. Descriptive statistical analyses (median and range) were performed with R (version 4.2.2)/R Studio(version 2022.12.0). QC materials for BF testing were purchased from Bio-Rad (Multilevel Qual, Levels 1 and 3; Hercules, CA, USA). Target concentrations are reported. Results Conclusion BF median (range) concentrations are similar or higher for bilirubin, creatinine, glucose, lipase, and UN. Serum QC material is likely appropriate for these assays in BFs. BF median(range) concentrations are lower for albumin, cholesterol, TP, and triglycerides suggesting that serum QC material may not be well suited for BFs. The need for extended dilutions may impact BF amylase and lipase.