Derivation and internal validation of the multivariate toxigenic C. difficile diarrhea model and risk score for emergency room and hospitalized patients with diarrhea.

Background: Many factors have been associated with the risk of toxigenic C. difficile diarrhea (TCdD). This study derived and internally validated a multivariate model for estimating the risk of TCdD in patients with diarrhea using readily available clinical factors. Methods: A random sample of 3,050 symptomatic emergency department or hospitalized patients undergoing testing for toxigenic C. difficile at a single teaching hospital between 2014 and 2018 was created. Unformed stool samples positive for both glutamate dehydrogenase antigen by enzyme immunoassay and tcdB gene by polymerase chain reaction were classified as TCdD positive. The TCdD Model was created using logistic regression and was modified to the TCdD Risk Score to facilitate its use. Results: 8.1% of patients were TCdD positive. TCdD risk increased with abdominal pain (adjusted odds ratio 1.3; 95% CI, 1.0-1.8), previous C. difficile diarrhea (2.5, 1.1-6.1), and prior antibiotic exposure, especially when sampled in the emergency department (4.2, 2.5-7.0) versus the hospital (1.7, 1.3-2.3). TCdD risk also increased when testing occurred earlier during the hospitalization encounter, when age and white cell count increased concurrently, and with decreased eosinophil count. In internal validation, the TCdD Model had moderate discrimination (optimism-corrected C-statistic 0.65, 0.62-0.68) and good calibration (optimism-corrected Integrated Calibration Index [ICI] 0.017, 0.001-0.022). Performance decreased slightly for the TCdD Risk Score (C-statistic 0.63, 0.62-0.63; ICI 0.038, 0.004-0.038). Conclusions: TCdD risk can be predicted using readily available clinical risk factors with modest accuracy.

Machine learning models identify predictive features of patient mortality across dementia types.

BACKGROUND: Dementia care is challenging due to the divergent trajectories in disease progression and outcomes. Predictive models are needed to flag patients at risk of near-term mortality and identify factors contributing to mortality risk across different dementia types. METHODS: Here, we developed machine-learning models predicting dementia patient mortality at four different survival thresholds using a dataset of 45,275 unique participants and 163,782 visit records from the U.S. National Alzheimer's Coordinating Center (NACC). We built multi-factorial XGBoost models using a small set of mortality predictors and conducted stratified analyses with dementiatype-specific models. RESULTS: Our models achieved an area under the receiver operating characteristic curve (AUC-ROC) of over 0.82 utilizing nine parsimonious features for all 1-, 3-, 5-, and 10-year thresholds. The trained models mainly consisted of dementia-related predictors such as specific neuropsychological tests and were minimally affected by other age-related causes of death, e.g., stroke and cardiovascular conditions. Notably, stratified analyses revealed shared and distinct predictors of mortality across eight dementia types. Unsupervised clustering of mortality predictors grouped vascular dementia with depression and Lewy body dementia with frontotemporal lobar dementia. CONCLUSIONS: This study demonstrates the feasibility of flagging dementia patients at risk of mortality for personalized clinical management. Parsimonious machine-learning models can be used to predict dementia patient mortality with a limited set of clinical features, and dementiatype-specific models can be applied to heterogeneous dementia patient populations.

Dementia has emerged as a major cause of death in societies with increasingly aging populations. However, predicting the exact timing of death in dementia cases is challenging, due to variations in the gradual process where cognitive decline interferes with the body's normal functions. In our study, we build machine-learning models to predict whether a patient diagnosed with dementia will survive or die within 1, 3, 5, or 10 years. We found that the prediction models can work well across patients from different parts of the US and across patients with different types of dementia. The key predictive factor was the information that is already used to diagnose and stage dementia, such as the results of memory tests. Interestingly, broader risk factors related to other causes of death, such as heart conditions, were less significant for predicting death in dementia patients. The ability of these models to identify dementia patients at a heightened risk of mortality could aid clinical practices, potentially allowing for earlier interventions and tailored treatment strategies to improve patient outcomes.