Establishment of a nutritional risk prediction model for elderly patients with cognitive impairment
10.3969/j.issn.1002-0152.2024.12.002
- VernacularTitle:老年认知功能障碍患者营养风险预测模型的建立
- Author:
Qian WANG
1
;
Haijun MIAO
1
;
Wei CHEN
1
;
Xiaohui ZHOU
1
Author Information
1. 新疆医科大学第一附属医院老年病科(乌鲁木齐 830054)
- Publication Type:Journal Article
- Keywords:
CONUT;
GNRI;
NRS2002;
Nutritional screening;
Cognitive impairment;
Prediction model;
Com-bined diagnostic
- From:
Chinese Journal of Nervous and Mental Diseases
2024;50(12):711-718
- CountryChina
- Language:Chinese
-
Abstract:
Objective Nutritional screening tools were used to assess the nutritional status of patients with cognitive impairment during hospitalization,and to explore the application value of objective nutrition screening tools in the nutritional evaluation of such patients.Methods Selected elderly people hospitalized in geriatrics from January 2019 to December 2021,a total of 340 elderly patients with cognitive impairment were enrolled using diagnostic test methods.The nutritional risk screening 2002(NRS2002)was selected as the gold standard,including 173 cases in the nutritional risk group and 167 cases in the non-nutrition risk group.The risk factors of the two groups were analyzed,the consistency between the controlling nutritional status score(CONUT),the geriatric nutritional risk index(GNRI)and NRS2002 was analyzed,and the predictive model was constructed by using logistic regression combined with ROC curve to evaluate the diagnostic efficacy of the model.Results The incidence of nutritional risk was relatively high in hospitalized elderly patients with cognitive impairment.The detection rates of all the three screening scales were above 50%.Body mass index(BMI)was an independent protective factor(P<0.05)while age was an independent risk factor(P<0.05)for elderly patients with cognitive impairment.The consistency between CONUT and NRS2002 was poor,while the consistency between GNRI and NRS2002 was moderate.The CONUT,GNRI,body mass index(BMI),serum sodium and high density lipoprotein(HDL)models were selected.The area under the curve of the combined diagnostic model could reach 0.905(95%CI:0.869-0.934),with a sensitivity of 83.8%,a specificity of 88.0%,and a Yoden index of 0.718.Conclusion CONUT,GNRI,BMI,serum sodium and HDL are independently associated with nutritional risk,and all have a certain independent predictive power.The diagnostic efficacy was optimal in the joint assessment.
