Food consumption in school-age children : a new web-based recall for dietary assessment in Portugal

Tese de doutoramento, Doenças Metabólicas e Comportamento Alimentar (Ciências e Tecnologias da Saúde) (Nutrição), Universidade de Lisboa, Faculdade de Medicina, 2016Accurate dietary assessment is critical for monitoring the nutritional status of children, examining associations between diet and health, and identifying dietary intake patterns and eating behaviours. The study of diets of children poses methodological problems relating to the accuracy of assessment. The 24-hour Dietary Recall (24-h DR) is among the most accurate methods to estimate total energy and nutrient intakes in school-age children. Traditional 24-h DR is expensive and impractical for large-scale studies. The application of technology to automate the more accurate 24-h DR is being conducted in multiple countries across the world, making them less expensive and easier to use. The objective of this study is to develop, validate and test a new web-based recall for dietary assessment in Portuguese school-age children: the Portuguese self-administered computerised 24-h DR (PAC24). The PAC24 is a self-administered web-based 24-h DR based on multiple pass method directed to second-, third- or fourth-grade Portuguese children (7-10 years old). In PAC24, children are first questioned about food and drink consumption on the previous day. Food entry is done via free text search, supported by a spell check application. For the majority of foods, amount consumed is estimated by selecting the closest portion size, served and leftover, if any, among seven different digital images. Data about time, place, television watching and computer use are assigned to each eating occasion. The food composition code and weight (g or mL) of selected items are automatically allocated and stored. A database of 380 food items is available. Food, energy and nutrient information is linked to a database that contains essentially information about Portuguese food composition table. The development of PAC24 was based on literature review, 21 focus groups (FG) developed in seven primary schools of the seven main regions of Portugal and input from national and international researchers with experience in computer-dietary assessment among children. A prototype of PAC24 was tested and its content was validated through an expert meeting. Accuracy was determined by comparison of PAC24 with lunch observations (‘gold standard’) in two schools in Lisbon and Tagus Valley (LTV) region. A pilot study was conducted to check the feasibility of PAC24 with respect to procedures, methods and data processing in one primary school in LTV region on two non-consecutive days with 15 days apart. The food consumption, particularly total energy and nutrient (protein, total fat, carbohydrate, fibre, calcium, sodium, potassium) intakes were studied. Wilcoxon signed-rank tests were used to compare the food consumption between the first and second measurement occasion. A p-value 0.05). In summary, PAC24 is a cost-effective, intuitive and an engaging method for Portuguese school-age children. The PAC24 could be used to estimate dietary intake on a group level, accurately. This method will provide useful information for epidemiological studies on the links between diet and health and contribute to the improvement of public health policies at national level