Comparison of six nutritional screening tool‐based Global Leadership Initiative on Malnutrition criteria models for predicting activities of daily living in older patients with subacute stroke: A single‐center retrospective cohort study
Shinnosuke Ishida, Ryota Imai, Yu Kitaji, Satoshi Miyano, Yoshiyuki KawanoAbstract
Background
The Global Leadership Initiative on Malnutrition (GLIM) recommends a two‐step diagnostic process consisting of initial screening for malnutrition risk followed by diagnostic assessment. We evaluated the association between GLIM‐defined malnutrition identified using six alternative first‐step screening tools and Functional Independence Measure (FIM)‐motor scores at discharge in older patients with subacute stroke.
Methods
This retrospective cohort study involved patients aged ≥70 years who were admitted to a convalescent rehabilitation ward after subacute stroke. Malnutrition risk was assessed using six tools: the Mini Nutritional Assessment–Short Form (MNA‐SF), Malnutrition Universal Screening Tool (MUST), Nutritional Risk Screening 2002 (NRS‐2002), Geriatric Nutritional Risk Index (GNRI), Controlling Nutritional Status (CONUT), and Prognostic Nutritional Index (PNI). Malnutrition was diagnosed based on the GLIM criteria. The outcome was the FIM‐motor score at discharge. Associations between GLIM‐defined malnutrition based on each screening tool and FIM‐motor scores were examined using hierarchical multiple regression analysis.
Results
Among the 152 patients included in the analysis, the prevalence of GLIM‐defined malnutrition ranged from 28.3% (PNI) to 40.8% (NRS‐2002). Malnutrition, as defined by GLIM and based on MNA‐SF, MUST, NRS‐2002, and GNRI, was significantly associated with lower FIM‐motor scores at discharge. The model incorporating MUST revealed the strongest association with FIM‐motor scores at discharge ( β = –0.11, P = 0.007, Δ R 2 = 0.012).
Conclusion
GLIM‐defined malnutrition was negatively associated with FIM‐motor scores at discharge in older patients with subacute stroke. The GLIM model incorporating MUST may be particularly useful for identifying malnutrition associated with FIM‐motor scores at discharge.