Development and validation of a nomogram for predicting the risk of nursing home-acquired pneumonia.

ABSTRACT

OBJECTIVE: Nursing home-acquired pneumonia (NHAP) is one of the most frequent infections in nursing homes, with a difficult diagnosis, poor prognosis, and high mortality. The present study was performed to develop and validate a nomogram to predict the risk of NHAP in nursing homes. MATERIALS AND METHODS: Based on a literature review and clinician's recommendations, we identified and collected the possible factors affecting the occurrence of NHAP. Based on the above factors, a retrospective observational study of 620 nursing home residents' medical records was performed from September 2016 to September 2021. Significant risk factors for NHAP were identified by univariate and multivariate analysis successively. A nomogram was constructed based on the binary logistic regression models to visualize the prediction model. The model's performance was determined by the concordance index (C-index), and the prediction accuracy was evaluated using a calibration curve. Clinical effectiveness was evaluated by decision curve analysis (DCA). RESULTS: Finally, 12 independent risk factors were identified and assembled into the nomogram. The nomogram had a C-index of 0.958 (95% confidence interval 0.943-0.972). The area under the receiver operating characteristic curve (AUC-ROC) value of the nomogram was 0.958 (p<0.05), and the calibration plot showed good goodness of fit. The decision curve analysis and clinical impact curve showed good clinical usefulness of the nomogram. CONCLUSIONS: A nomogram for the early prediction of NHAP, which is easy for nurses to perform in nursing homes, was successfully constructed and validated, and it had a good predictive performance.