AbstractsMedical & Health Science

Development, validation and reliability of a short food frequency questionnaire that measures sugar intake in Māori living in Gisborne, New Zealand

by Hannah Esther Walter




Institution: University of Otago
Department:
Year: 0
Keywords: Dietary Assessment; Food frequency questionnaires; Sugar; Validity; Reliability; Māori adults; Dietary; Assessment; Food; Frequency; Questionnaires; Sugar; Validity; Reliability; Maori; Adults; Dietary Assessment; Food frequency questionnaires; Sugar; Validity; Reliability; Māori adult
Record ID: 1314678
Full text PDF: http://hdl.handle.net/10523/5336


Abstract

Background: Worldwide, obesity, diabetes, cardiovascular and metabolic disease epidemics have increased in parallel with increasing levels of sugar intake, suggesting the idea that sugar may be partially responsible. In New Zealand (NZ) these chronic health conditions are more prevalent in the indigenous Māori population than the general population. To ascertain whether high sugar intake, and fructose in particular, is associated with chronic disease risk in this population, there is a need develop a valid and reliable dietary assessment instrument, that is easy to administer and comprehend, to measure sugary food and beverage intakes. Objective: To develop, pretest and validate culturally appropriate self-administered, semi-quantitative food frequency questionnaire (FFQ), designed to measure and rank individuals usual sugars intakes over the previous month in Māori adults. Design: An existing closed-end FFQ designed to identify people with high sugars intakes in a previous research study was adapted and pretested in both closed and open-ended formats in a sample of ten Māori adults living in Gisborne using cognitive interviewing techniques. Important sugar-containing food and beverage sources were identified from various sources including published FFQs, and the New Zealand Adult Nutrition Survey 2008/2009 as well as an environmental audit of local food outlets and supermarkets. Modifications made as a result of pretesting included adding additional foods and photographs, and rephrasing of questions. The final FFQ consisted of 33 items, with participants preferring the open-ended format. The FFQ was validated by comparing sugars estimates from the FFQ with estimates from three repeated 24hr diet recalls conducted over the preceding month in a convenience sample of 35 Māori adults, aged 20-65 years. Reliability of the FFQ was assessed by comparing sugars estimates from two administrations of the FFQ conducted 4 weeks apart. Results: There was no significant difference between the mean intake from the FFQ and 24hr recalls for fructose, glucose and total sugar. Mean sucrose intake was significantly different (P=0.03) between the two instruments. Bland-Altman analyses for agreement between the FFQ and 2hr recalls ranged from 101 for fructose to 124 for sucrose. The limits of agreement (LoA) were wide; for fructose the LoA ranged from 32% - 318%. Classification into the same or adjacent quartile was 97% for fructose and 100% for all othersugars categories with weighted Kappa scores ranging from 0.47 for sucrose to 0.58 for glucose. Sugars estimates from both instruments were strongly correlated (Spearmans coefficients all ≥0.69). Intra-class correlations for sugars estimates obtained from the two FFQ administrations were all >0.8. Conclusion: The Kai with Māori FFQ was found to be a reliable instrument that can adequately rank individuals according to their sugars intake as well as estimate group mean intakes. The FFQ provides a simple, efficient and cost effective method to measure sugar intake within the…