目的

评价全球领导人发起的营养不良诊断标准共识（GLIM）在老年住院患者营养不良诊断中的效率，并筛选用于“营养筛查-评定-诊断”临床路径的最佳营养评估工具组合。

方法

采用横断面研究设计，分别以营养风险筛查2002（NRS-2002）、微型营养评定量表简表（MNA-SF）、老年营养风险指数（GNRI）、主观全面营养评定（SGA）作为营养不良的判定标准，进行GLIM诊断试验。以Kappa （κ）作为信度评价指标，以灵敏度、特异度、阳性预测值、阴性预测值、约登指数（γ）作为效度评价指标。

结果

分别以NRS-2002、MNA-SF、GNRI、SGA、GLIM为标准进行评定，营养不良的发生率分别为75.64%、47.12%、55.45%、74.68%、51.28%。其中GLIM与MNA-SF和SGA在诊断老年住院患者营养不良上有较高的一致性（κ=0.824、 0.824，γ=0.79、0.83），而GLIM与NRS2002和GNRI在诊断老年住院患者营养不良上的一致性较弱（κ=0.480、0.344，γ=0.48、0.34）。

结论

住院老年患者中营养不良或存在营养不良风险者的占比较高，可先使用NRS2002和GNRI进行营养风险筛查，再使用GLIM进行营养评估和营养不良诊断。

Objective

To evaluate the efficiency of the Global Leadership Initiative on Malnutrition （GLIM） consensus criteria for diagnosing malnutrition in elderly inpatients， and to identify the optimal combination of nutritional assessment tools for the "nutritional screening-assessmentdiagnosis" clinical pathway.

Methods

A cross-sectional study was performed to compare GLIM with the nutritional risk screening 2002 （NRS2002）， the mini-nutritional assessment short-form （MNA-SF），the geriatric nutritional risk index （GNRI）， and the subjective global assessment （SGA）. The reliability of GLIM was evaluated by the Kappa （κ）. Sensitivity， specificity， the positive predictive value， the negative predictive value and Youden's index （γ） were used to evaluate GLIM's validity.

Results

Using NRS-2002， MNA-SF， GNRI， SGA， and GLIM as the diagnostic criteria for malnutrition， the incidence rates of malnutrition were 75.64%， 47.12%， 55.45%， 74.68%， and 51.28%， respectively. Among these， GLIM showed a high level of agreement with MNA-SF and SGA （κ=0.824， 0.824； γ=0.79， 0.83）， while the agreement between GLIM and NRS2002 as well as GNRI was relatively weak （κ=0.480， 0.344； γ=0.48，0.34） .

Conclusion

The prevalence of malnutrition or the risk of malnutrition among hospitalized elderly patients is relatively high. It is recommended to use NRS2002 and GNRI for nutritional risk screening， followed by GLIM for nutritional assessment and malnutrition diagnosis.