Background: Older people are more likely to develop nutritional problems and timely diagnosis of malnutrition is crucial to prevent hazardous consequences following poor nutrition.
Aims: To evaluate the efficacy of Geriatric Nutritional Risk Index (GNRI) to assess nutritional status among non-hospitalized elderly, compared to mini nutritional assessment (MNA) among Iranian seniors.
Methods: One hundred and sixty-four subjects, aged ≥ 65 years old were recruited to our cross-sectional study from various districts of Tabriz (Tabriz, Iran). Anthropometric and biochemical measurements were performed, short- and long-form MNAs and GNRI were assessed in our study subjects. Sensitivity, specificity and predictive values of the three indices, agreement between them, and their correlation with anthropometric and biochemical parameters were evaluated. Receiver-operating characteristic (ROC) curve analysis was performed to determine the optimal cut-off point for GNRI in our study population.
Results: GNRI had lower sensitivity (50, 57%), but optimal specificity (94, 93%) and lower negative predictive value (NPV; 68, 71%) compared to MNA-LF and MNA-SF, respectively. We found a moderate agreement between GNRI and MNA-SF (K = 0.52) and MNA-LF (K = 0.46) scores. Significant correlations were observed between re-categorized MNAs as well as GNRI scores, and age, weight, MAC, CC, WC, albumin, and pre-albumin. The cut-off point of 110.33 was obtained for GNRI, according to the ROC curve.
Conclusions: Although GNRI may not be an efficient tool for screening malnutrition due to its lower sensitivity, it is moderately correlated with MNAs and also more useful when limited funding needs to target the truly malnourished seniors.
Keywords: Elderly; Geriatric Nutritional Risk Index (GNRI); Mini nutritional assessment (MNA); Sensitivity; Specificity.