{"title":"预测认知障碍营养风险的Nomogram。","authors":"Yuhang Chen, Junlin Diao, Xuezhuang Ren, Chunxiang Wei, Xue Zhou","doi":"10.1177/25424823241309262","DOIUrl":null,"url":null,"abstract":"<p><strong>Background: </strong>Cognitive impairment patients are prone to malnutrition, which further promotes cognitive decline. Cognitive impairment patients are unable to accurately answer subjective questions in the nutrition screening scale. Therefore, it is crucial to establish a nutritional risk prediction model using objective evaluation indicators to evaluate the nutritional status of cognitive impairment patients during hospitalization.</p><p><strong>Objective: </strong>To develop a nomogram for prediction of the nutritional risk in cognitive impairment patients.</p><p><strong>Methods: </strong>The least absolute shrinkage and selection operator (LASSO) was used for regression analysis, and predictive factors were selected based on 10-fold cross validation. Then, using the selected predictive factors, multivariable logistic regression analysis was performed to obtain the final clinical prediction model. Moreover, the performance of the model was evaluated from receiver operating characteristic curve, calibration curve, and decision curve analysis. Further assessment was conducted through internal validation.</p><p><strong>Results: </strong>Six predictive factors were selected from 20 variables through LASSO, including body mass index, age, triglyceride, taking cognitive-improving drugs, controlling nutritional status, and geriatric nutritional risk index. The area under the receiver operating characteristic curve of the training cohort was 0.91, while the validation cohort was 0.88, indicating that the model constructed with 6 predictors had moderate predictive ability. The decision curve analysis showed that the threshold range for both groups was 0.00-0.80, with the highest net benefit 0.76 for training cohort, while 0.77 for validation cohort.</p><p><strong>Conclusions: </strong>Introducing six predictive factors, the risk nomogram is useful for predicting nutritional risk of cognitive impairment.</p>","PeriodicalId":73594,"journal":{"name":"Journal of Alzheimer's disease reports","volume":"9 ","pages":"25424823241309262"},"PeriodicalIF":2.8000,"publicationDate":"2025-01-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11864246/pdf/","citationCount":"0","resultStr":"{\"title\":\"Nomogram for predicting nutritional risk of cognitive impairment.\",\"authors\":\"Yuhang Chen, Junlin Diao, Xuezhuang Ren, Chunxiang Wei, Xue Zhou\",\"doi\":\"10.1177/25424823241309262\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<p><strong>Background: </strong>Cognitive impairment patients are prone to malnutrition, which further promotes cognitive decline. Cognitive impairment patients are unable to accurately answer subjective questions in the nutrition screening scale. Therefore, it is crucial to establish a nutritional risk prediction model using objective evaluation indicators to evaluate the nutritional status of cognitive impairment patients during hospitalization.</p><p><strong>Objective: </strong>To develop a nomogram for prediction of the nutritional risk in cognitive impairment patients.</p><p><strong>Methods: </strong>The least absolute shrinkage and selection operator (LASSO) was used for regression analysis, and predictive factors were selected based on 10-fold cross validation. Then, using the selected predictive factors, multivariable logistic regression analysis was performed to obtain the final clinical prediction model. Moreover, the performance of the model was evaluated from receiver operating characteristic curve, calibration curve, and decision curve analysis. Further assessment was conducted through internal validation.</p><p><strong>Results: </strong>Six predictive factors were selected from 20 variables through LASSO, including body mass index, age, triglyceride, taking cognitive-improving drugs, controlling nutritional status, and geriatric nutritional risk index. The area under the receiver operating characteristic curve of the training cohort was 0.91, while the validation cohort was 0.88, indicating that the model constructed with 6 predictors had moderate predictive ability. The decision curve analysis showed that the threshold range for both groups was 0.00-0.80, with the highest net benefit 0.76 for training cohort, while 0.77 for validation cohort.</p><p><strong>Conclusions: </strong>Introducing six predictive factors, the risk nomogram is useful for predicting nutritional risk of cognitive impairment.</p>\",\"PeriodicalId\":73594,\"journal\":{\"name\":\"Journal of Alzheimer's disease reports\",\"volume\":\"9 \",\"pages\":\"25424823241309262\"},\"PeriodicalIF\":2.8000,\"publicationDate\":\"2025-01-13\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11864246/pdf/\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Journal of Alzheimer's disease reports\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1177/25424823241309262\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"2025/1/1 0:00:00\",\"PubModel\":\"eCollection\",\"JCR\":\"Q2\",\"JCRName\":\"NEUROSCIENCES\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Journal of Alzheimer's disease reports","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1177/25424823241309262","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"2025/1/1 0:00:00","PubModel":"eCollection","JCR":"Q2","JCRName":"NEUROSCIENCES","Score":null,"Total":0}
Nomogram for predicting nutritional risk of cognitive impairment.
Background: Cognitive impairment patients are prone to malnutrition, which further promotes cognitive decline. Cognitive impairment patients are unable to accurately answer subjective questions in the nutrition screening scale. Therefore, it is crucial to establish a nutritional risk prediction model using objective evaluation indicators to evaluate the nutritional status of cognitive impairment patients during hospitalization.
Objective: To develop a nomogram for prediction of the nutritional risk in cognitive impairment patients.
Methods: The least absolute shrinkage and selection operator (LASSO) was used for regression analysis, and predictive factors were selected based on 10-fold cross validation. Then, using the selected predictive factors, multivariable logistic regression analysis was performed to obtain the final clinical prediction model. Moreover, the performance of the model was evaluated from receiver operating characteristic curve, calibration curve, and decision curve analysis. Further assessment was conducted through internal validation.
Results: Six predictive factors were selected from 20 variables through LASSO, including body mass index, age, triglyceride, taking cognitive-improving drugs, controlling nutritional status, and geriatric nutritional risk index. The area under the receiver operating characteristic curve of the training cohort was 0.91, while the validation cohort was 0.88, indicating that the model constructed with 6 predictors had moderate predictive ability. The decision curve analysis showed that the threshold range for both groups was 0.00-0.80, with the highest net benefit 0.76 for training cohort, while 0.77 for validation cohort.
Conclusions: Introducing six predictive factors, the risk nomogram is useful for predicting nutritional risk of cognitive impairment.