|
Abstract
|
BACKGROUND: Fall is a public health problem that cannot be ignored by elderly stroke patients, and rehabilitation care plays an important role in the rehabilitation process of elderly stroke patients.
OBJECTIVE: To investigate the prevention effect of fall risk in elderly stroke patients under the intelligent model of rehabilitation care.
METHODS: The general data of elderly patients who were diagnosed as stroke and admitted to our hospital between June 2021 and June 2022 were retrospectively analyzed, with exclusion like unclear clinical data or combined with other severe organ insufficiency. A total of 150 of them were selected for the study, and the patients were divided into a fall group and a non-fall group according to whether they had a fall or not. The factors associated with falls in stroke patients were analyzed univariately, and the rehabilitation care intelligence model of the predictive model of falls in stroke patients was established using multiple covariance ridge regression analysis to observe the predictive value of patients' risk of falling in the rehabilitation care intelligence model.
RESULTS: Results of multiple covariance ridge regression analysis to build the model showed age (P <.001), low MNA-SF score (P <.001), hypertension (P =.035), anaemia(P =.048), gout (P <.001), assistive devices (P =.002), visual impairment (P =.033), elevated ALB (P <.001), and elevated HGB (P <.001) as risk factors for falls in stroke patients. The diagnostic threshold for screening elderly stroke patients for falls based on risk factors was 0.272, with a sensitivity of 90.7%, specificity of 98.1% and an area under the ROC curve of 0.976 (P <.05), which was superior to other single indicators in terms of diagnostic value. The calibration of the prediction model, based on the Hosmer and Lemeshow test of goodness of fit, showed P = 1.14, indicating a high calibration of the prediction model.
CONCLUSION: There are many risk factors for falls in stroke elderly patients, such as low MNA-SF score, gout, elevated ALB, and elevated HGB. Building a rehabilitation nursing intelligent model based on the above inducement factors can reduce the risk of patients falling to a certain extent, and the prediction model has a high degree of calibration. Therefore, a simple and standardized intelligent rehabilitation nursing model for stroke patients in the early stage can effectively prevent the occurrence of falls.
Language: en |