Abstract
OBJECTIVE: To develop and internally validate a Free Sugars Screener (FSS) for Australian children aged 2 and 5 years. DESIGN: Using data collected from a ninety-nine-item (2-year-olds) and ninety-eight-item (5-year-olds) FFQ in the Study of Mothers' and Infants' Life Events affecting oral health (SMILE-FFQ), a regression-based prediction modelling approach was employed to identify a subset of items that accurately estimate total free sugars intake (FSI). The predictors were grams of free sugars (FSg) for individual items in the SMILE-FFQ and child's age and sex. The outcome variable was total FSI per person. To internally validate the SMILE-FSS items, the estimated FSg was converted to percent energy from free sugars (%EFS) for comparison to the WHO free sugars guideline categories (< 5 %, 5-< 10 % and ≥ 10 %EFS) using cross-classification analysis. SETTING: Australia. PARTICIPANTS: 858 and 652 2- and 5-year-old children, respectively, with complete dietary (< 5 % missing) and sociodemographic data. RESULTS: Twenty-two and twenty-six items were important in predicting FSI at 2 and 5 years, respectively. Items were similar between ages with more discretionary beverage items (e.g. sugar-sweetened beverages) at 5 years. %EFS was overestimated by 4·4 % and 2·6 %. Most children (75 % and 82 %) were categorised into the same WHO free sugars category with most (87 % and 95 %) correctly identified as having < 10 %EFS in line with the WHO recommendation. CONCLUSIONS: The SMILE-FSS has good internal validity and can be used in research and practice to estimate young Australian children's FSI and compare to the WHO free sugars guidelines to identify those 'at risk'.