Conectar
by Ikgyu Shin, Nilay Bhatt, Alaa Alashi, Keervani Kandala, Karthik Murugiah
ObjectivesTo develop and compare prediction models for 30-day and 1-year mortality in Heart failure with preserved ejection fraction (HFpEF) using EHR data, utilizing both traditional and machine learning (ML) techniques.
BackgroundHFpEF represents 1 in 2 heart failure patients. Predictive models in HFpEF, specifically those derived from electronic health record (EHR) data, are less established.
MethodsUsing MIMIC-IV EHR data from 2008−2019, patients aged ≥ 18 years admitted with a primary diagnosis of HFpEF were identified using ICD-9 and 10 codes. Demographics, vital signs, prior diagnoses, and lab data were extracted. Data was partitioned into 80% training, 20% test sets. Prediction models from seven model classes (Support Vector Classifier (SVC), Logistic Regression, Lasso Regression, Elastic Net, Random Forest, Histogram-based Gradient Boosting Classifier (HGBC), and eXtreme Gradient Boosting (XGBoost)) were developed using various imputation and oversampling techniques with 5-fold cross-validation. Model performance was compared using several metrics, and individual feature importance assessed using SHapley Additive exPlanations (SHAP) analysis.
ResultsAmong 3,235 hospitalizations for HFpEF, 30-day mortality was 6.3%, and 1- year mortality was 29.2%. Logistic regression performed well for 30-day mortality (Area Under the Receiver operating characteristic curve (AUC) 0.83), whereas Random Forest (AUC 0.79) and HGBC (AUC 0.78) for 1-year mortality. Age and NT-proBNP were the strongest predictors in SHAP analyses for both outcomes.
ConclusionModels derived from EHR data can predict mortality after HFpEF hospitalization with comparable performance to models derived from registry or trial data, highlighting the potential for clinical implementation.
Neurogranin (Ng) has a role in synaptic plasticity and is considered a biomarker of synaptic dysfunction, a process hypothesised to be important in delirium. Few studies examining Ng in delirium exist, with mixed findings. This study aimed to investigate associations between cerebrospinal fluid (CSF) Ng concentrations and delirium in acutely admitted hip fracture patients.
Cross-sectional study.
Acutely admitted orthopaedic patients with hip fracture recruited from four participating hospitals in eastern Norway, representing secondary and tertiary care settings.
This study included 392 hip fracture patients. All admitted hip fracture patients operated in spinal anaesthesia were, regardless of age, considered for inclusion.
An in-house ELISA was used to measure CSF Ng concentration in patients acutely admitted with a hip fracture (n=392). Delirium status was evaluated daily according to The Diagnostic and Statistical Manual of Mental Disorders, Fifth Editions criteria independently by two experienced geriatricians. A value
180 patients (46 %) developed delirium and 70% of these had dementia. CSF Ng concentration did not differ significantly between those with and without delirium (176 pg/mL vs 164 pg/mL), with an estimated difference in medians of 12 (95% CI –5.8 to 29.8), p=0.185. Analyses adjusted for age, gender and dementia status did not show a statistically significant difference in Ng concentrations between the patients.
We did not find an association between delirium and CSF concentrations of Ng. This could imply that synaptic dysfunction and degeneration, involving Ng, are not key processes in the development of delirium. Further studies on other synaptic proteins are warranted to better explore synaptic dysfunction’s potential role in the pathophysiology of delirium.
Non-specific symptoms of testosterone deficiency (TD) and lack of awareness impact diagnosis and appropriate treatment. This study aimed to characterise the awareness of key symptoms of TD in community-dwelling men and contextualise this against the reported prevalence of these symptoms.
Cross-sectional survey comprising 54 questions (including assessment of symptoms as per the qADAM questionnaire and where relevant, men’s experiences while on TD treatment). The survey was distributed through online media channels, Prolific and academic networks.
Community-dwelling men in the UK.
Associations between age, participant demographics and a ‘positive’ qADAM score were assessed using logistic regression. A positive qADAM score was defined as self-rated ‘poor’ or ‘terrible’ libido or erection strength or rating 3 of the other questionnaire domains as ‘poor’ or ‘terrible’.
Of 973 men, 49% indicated high likelihood of TD using qADAM scores—5% were formally diagnosed. Men over 50 years of age had 1.54–2.0 times higher odds of TD compared with men aged
Almost half of the responders exhibited a burden of TD-associated symptoms, but under 5% had a formal diagnosis. These findings suggest significant gaps between symptom awareness and access to treatment options.
FreshRSS 