Validity assessment of the nutrition screening initiative checklist in older adults

Summary Objective To ascertain whether the NSI (Nutrition Screening Initiative Checklist) is a valid screening instrument for diagnosing nutritional risk in an older population from a medium-size city in the south of Brazil. Methods The study population comprised individuals aged 60 years or older selected from a population-based cross-sectional survey. Data collection was carried out at two different timepoints: first, the NSI (instrument under test) was applied and later a dietary recall (gold standard) based on the previous day. Validity analyses were performed based on sensitivity and specificity as well as the Receiver Operating Characteristic Curve. Results A total of 183 older adults were studied. Agreement between the dietary recall and the NSI for positive and negative results was 54.6%.The sensitivity and specificity values were low for all cut-off points of the instrument. The area under the curve was 0.52 (95%CI: 0.44–0.62) for the cut-off point ≥6. Conclusion Based on sensitivity and specificity values, the NSI proved ineffective for application in the population studied.