Household Food Sufficiency Status and the Qualityof Children's Diets: An Analysis from the Continuing Survey of Food Intakes by Individuals 1994-1996, 1998

The relationship between food sufficiency and diet quality was explored among children 2-8 years of age living in households ≤185% poverty with 2-days of dietary recall data from 1994-96, 1998 Continuing Survey of Food Intakes by Individuals. Diet quality was assessed using measures of both adequacy and variety. Diet adequacy was measured by degree of adherence to age-specific daily serving recommendations for the 5 Food Guide Pyramid food groups and by intake of discretionary fat (grams) and added sugars (teaspoons). Variety was measured using the Healthy Eating Index Variety Score ( overall variety), the Dietary Diversity Score (among food group variety) and the Sub-Group Contribution Score (within food group variety).

When testing measures of adequacy, this study found that household food sufficiency status did not affect the ability to adhere to the serving recommendations for the major food groups of the Food Guide Pyramid and did not influence discretionary fat intake among low-income children ages 2-8 years. It did, however, affect consumption of added sugars in children 4-8 years of age. Furthermore, although the younger 2-3 year old low-income children seemed to eat a better diet than their 4-8 year old counterparts, both groups of children on average consumed diets that did not conform to the Food Guide Pyramid recommendations.

This study also found that household food sufficiency status did not affect the three measures of variety used. However, participation in the WIC Program was a significant predictor of overall variety (2-3 year olds) and among food group variety (2-8 year olds). Variety within food groups, as measured by the Sub-Group Contribution Scores, lent no support to the concepts set forth in the qualitative research regarding hunger and its affect on the eating patterns of children. Some trends between age groups and food sufficiency status were noted. However, these trends were not statistically significant when tested while controlling for other variables that may affect eating patterns.