Development and verification of a prediction model for outcomes of elderly patients with nursing home-acquired pneumonia.

ABSTRACT

BACKGROUND: Among all infections in nursing homes, pneumonia has the highest mortality. Nurses have a 24-h relationship with patients and have a key role in identifying and preventing adverse outcomes. However, tools to engage nurses in pneumonia patient outcomes evaluation have not occurred. PURPOSE: This study aimed to develop and validate a prediction model to predict the outcome of elderly patients with nursing home-acquired pneumonia (NHAP). METHODOLOGY: A retrospective observational study was conducted with 219 elderly NHAP patients. Baseline characteristics, health history, and treatment/nursing status were collected. Variables for constructing nomograms were screened by univariate and multivariate analysis. The nomogram model was evaluated using the concordance index (C-index), decision curve analysis (DCA) curves, and receiver operating characteristic (ROC) curves. RESULTS: 9 independent risk factors were identified and assembled into the nomogram. The nomogram exhibited reasonably accurate discrimination (area under the receiver operating characteristic curve (AUC-ROC) 0.931, P < 0.05) and calibration (C-index 0.931, 95 % CI 0.898-0.964) in the validation cohort. DCA and clinical impact curves demonstrated that the nomogram was clinically beneficial. CONCLUSIONS: A visualization nomogram model was successfully established for predicting the outcome of the NHAP elderly patients. The model has extremely high reliability, extremely high predictive ability, and good clinical applicability.