Abstract

Objectives: The aim of this systematic review was to summarize the validity of nutritional screening tools to detect the risk of malnutrition in community-dwelling older adults.

Design: A systematic review and meta-analysis. The protocol for this systematic review was registered in the PROSPERO database (CRD42017072703).

Setting and participants: A literature search was performed in PubMed, EMBASE, CINAHL, and Cochrane using the combined terms "malnutrition," "aged," "community-dwelling," and "screening." The time frame of the literature reviewed was from January 1, 2001, to May 18, 2018. Older community-dwellers were defined as follows: individuals with a mean/median age of >65 years who were community-dwellers or attended hospital outpatient clinics and day hospitals. All nutritional screening tools that were validated in community-dwelling older adults against a reference standard to detect the risk of malnutrition, or with malnutrition, were included.

Measures: Meta-analyses were performed on the diagnostic accuracy of identified nutritional screening tools validated against the Mini Nutritional Assessment-Long Form (MNA-LF). The symmetric hierarchical summary receiver operating characteristic models were used to estimate test performance.

Results: Of 7713 articles, 35 articles were included in the systematic review, and 9 articles were included in the meta-analysis. Seventeen nutritional screening tools and 10 reference standards were identified. The meta-analyses showed average sensitivities and specificities of 0.95 (95% confidence interval [CI] 0.75-0.99) and 0.95 (95% CI 0.85-0.99) for the Mini Nutritional Assessment-Short Form (MNA-SF; cutoff point ≤11), 0.85 (95% CI 0.80-0.89) and 0.87 (95% CI 0.86-0.89) for the MNA-SF-V1 (MNA-SF using body mass index, cutoff point ≤11), 0.85 (95% CI 0.77-0.89) and 0.84 (95% CI 0.79-0.87) for the MNA-SF-V2 (MNA-SF using calf circumference instead of body mass, cutoff point ≤11), respectively, using MNA-LF as the reference standard.

Conclusions and implications: The MNA-SF, MNA-SF-V1, and MNA-SF-V2 showed good sensitivity and specificity to detect community-dwelling older adults at risk of malnutrition validated against the MNA-LF. Clinicians should consider the use of the cutoff point ≤11 on the MNA-SF, MNA-SF-V1, and MNA-SF-V2 to identify community-dwelling older adults at risk of malnutrition.