Investigate the influencing factors of stroke-associated pneumonia (SAP) in patients with post-stroke dysphagia, and to develop a risk prediction model for SAP, as well as verify its predictive effectiveness.

Prospective continuous inclusion of patients with post-stroke dysphagia admitted to the neurology departments of three hospitals in Chongqing from November 2022 to May 2023. Investigated the general characteristics, clinical data, swallowing function, and oral health status of the patients. An elastic net method was applied to select predictive factors, followed by constructing a multiple-factor logistic regression model and validating it. The results were presented using nomogram to display the outcomes of the model.

A total of 211 patients were included, 56(21.80%)of whom developed SAP. The SAP prediction model included 5 items, including national institutes of health stroke scale(NIHSS) score, Water swallow test, salivation status, OHAT score and oral hygiene status. The likelihood ratio test results showed that the model was statistically significant (χ²=132.62, P<0.001)and the discrimination index was 0.960. The area under the ROC curve of this model was 0.93(95%CI: 0.87-0.93), with the optimal critical value of 0.10, accuracy of 83.87%, sensitivity of 84.62% and the specificity of 83.67%.

The constructed model can effectively predict the risk of stroke-associated pneumonia (SAP) in patients with post-stroke dysphagia. This has clinical significance in identifying high-risk patients early and developing personalized preventive strategies.