Send to

Choose Destination
Eur J Clin Nutr. 2019 Sep;73(9):1260-1269. doi: 10.1038/s41430-019-0411-3. Epub 2019 Feb 28.

SCREEN III: working towards a condensed screening tool to detect nutrition risk in community-dwelling older adults using CLSA data.

Author information

Department of Kinesiology, University of Waterloo, Waterloo, ON, Canada.
School of Public Health and Health Systems, University of Waterloo, Waterloo, ON, Canada.
Department of Kinesiology, University of Waterloo, Waterloo, ON, Canada.
Schlegel-UW Research Institute for Aging, Waterloo, ON, Canada.



Screening for nutrition risk in community-dwelling older adults increases the likelihood of early intervention to improve nutritional status, with short screening tools preferred. SCREEN-II-AB is a valid 8-item tool. The current study determines whether SCREEN-III, a proposed 3-item version, adequately classifies nutrition risk in comparison.


Baseline data from the Canadian Longitudinal Study on Aging were used. Seventy-two percent (n = 24,456) of eligible participants (>55 years, complete SCREEN-II-AB) were included. Sensitivity and specificity of various SCREEN-III values compared with SCREEN-II-AB risk determined a nutrition risk cut-point and the proportion misclassified (False[-]) was calculated. Construct validity was tested against a composite variable summarizing outcomes associated with nutrition risk (e.g., self-reported health, hospitalization) using logistic regression adjusted for individual factors (e.g., marital status).


A SCREEN-III cut-point of <22 performed best on sensitivity (0.83 [95% CI = 0.82, 0.84]) and specificity (0.73 [95% CI = 0.72, 0.74]) compared to SCREEN-II-AB (Cramer's V = 0.53). Of those at-risk using SCREEN-II-AB, 16.7% were misclassified as False(-) by SCREEN-III. The False(-) group did not differ significantly from the True(-) group. Based on SCREEN-III, 45.3% of individuals were at nutrition risk, 44% of whom reported the outcome composite. SCREEN-III nutrition risk was associated with greater odds of the outcome composite compared to those not at-risk (OR = 1.40, 95% CI = 1.33, 1.48, P < 0.0001).


The proposed version of SCREEN-III demonstrated construct validity, but misclassification of risk may be problematic; further validation of a 3-item version is recommended.


Supplemental Content

Full text links

Icon for Nature Publishing Group
Loading ...
Support Center