Assessing dietary intake in primary school children
The foods we eat in childhood impact on health in adult life. With the increasing incidence of diet related diseases such as non insulin dependent diabetes and cancer it is important that food intakes are monitored. Further in order to assess the effectiveness of health promotion initiatives methods of assessing intake are required which are both accurate and sensitive enough to detect changes in diet. If nutrient intakes are of interest these methods must include a measure or estimate of the amount of food consumed. Weighing foods imposes a large burden on the subject, may not be practical or possible in some sub-sections of populations e. g. children, and often results in underreporting. The purpose of this work was to develop methods for assessing dietary intake in 4 to 11 year olds; to assess the relative validity of these methods; to utilise the methods to assess the effectiveness of a dietary intervention and to assess the validity of current methods of assessing portion size for use with children. Two methods of assessing dietary intake were developed and pilot tested with children aged 4 to 11 years old. A food record designed to measure frequency of fruit and vegetable intake and a food diary with interview using food photographs to measure nutrient intake. Following refinement the methods were used to assess the effectiveness of a fruit and vegetable intervention. In a further study the validity of adult food photographs and food models in estimating portion size with children was assessed in an interview where children were shown known weights of foods. The food record and food diary were successful in detecting changes in intake of fruit and vegetables as a result of the intervention. The food record was found to be difficult to complete and was accurate in measuring fruit and vegetable intakes only at the group level. Accuracy of chi ' ldren's estimates of portion size were poor, children significantly overestimated food portion sizes on average using both the food photographs and the food models. The precision of children's estimates of portion size was also poor with a large range of over- and underestimates of portion size using both the food models and the food photographs.