Development and evaluation of a stand-alone index for the assessment of small children's diet quality.

Objective: To construct and evaluate an independent Children's Index of Diet Quality (CIDQ).

Design: A food consumption questionnaire, which contained twenty-five multiple-item questions on eating and food intake, was formulated and evaluated against 7 d food records. Key questions that best reflected a healthy diet, defined in criteria set by the nutrient recommendations, were searched and validated by correlation and analyses of receiver-operating characteristic curves. Settings A cohort of a young population of South-West Finland.

Subjects: Participants (n 400) were 2-6-year-old children.

Results: Fifteen questions were identified to best depict the children's diet quality in reference to the recommendations. These questions were scored, summarized and further constructed into a three-class index (good, moderate and poor dietary quality) where higher scores depicted better diet quality. The CIDQ cut-off score of 14 points for good dietary quality had a sensitivity of 0.59 and a specificity of 0.82 and the cut-off score of 10 points, for at least moderate dietary quality, had a sensitivity of 0.77 and a specificity of 0.69. Higher index scores were related to higher dietary intakes of several vitamins, lower dietary intakes of SFA and cholesterol, and further with lower serum cholesterol and higher serum vitamin C concentrations.

Conclusions: The three-class food index was found to represent diet quality as defined in recommendations and evaluated against nutrient intakes from food diaries and biochemical markers. This self-standing index could provide an effective and low-burden method to obtain information about diet quality and guide future recommendations.