Adaptation, acceptability and feasibility of a Short Food Survey to assess the dietary intake of children during attendance at childcare

Objective: To (i) describe the adaptation of the Short Food Survey (SFS) for assessing the dietary intake of children (2-5 years) during attendance at Early Childhood Education and Care (SFS-ECEC); (ii) determine the acceptability and feasibility of the SFS-ECEC; and (iii) compare the SFS-ECEC to direct observations for assessing dietary intake of children in care. Design: The adapted forty-seven-item SFS-ECEC was completed by childcare educators to capture individual child's usual intake over the past month. Acceptability and feasibility were assessed via educator self-report and completion rates. Mean servings of food groups consumed in accordance with dietary guidelines reported in the SFS-ECEC were compared to those obtained by a single-day direct observation via visual estimation conducted by trained personnel. Mean differences, intra-class correlations, Bland-Altman plots, percentage agreement and Cohen's kappa were examined. Setting: Early Childhood Education and Care, NSW, Australia. Participants: Educators and children. Results: 213 (98 center dot 61 %) SFS-ECECs were returned. Acceptability was high with 86 center dot 54 % of educators reporting the tool as easy to understand. Mean differences in servings of food groups between the SFS-ECEC and direct observation were statistically significantly different for five out of six foods and ranged 0 center dot 08-1 center dot 07, with intra-class correlations ranging 0 center dot 00-0 center dot 21. Agreement between the methods in the classification of children meeting or not meeting dietary guidelines ranged 42 center dot 78-93 center dot 01 %, with Cohen's kappa ranging -0 center dot 03 to 0 center dot 14. Conclusions: The SFS-ECEC is acceptable and feasible for completion by childcare educators. While tool refinement and further validation is warranted, small mean differences suggest the tool may be useful in estimating group-level intakes.