Abstract
BACKGROUND: Cereals have been extensively advocated as the beneficial food group in terms of body weight management, but each staple cereal crop may contribute in different ways. Studies of the association between wheat availability and risk of obesity are controversial. This study aimed to test the global and regional association between wheat availability as reported by FAO and obesity prevalence at a population level. FAO does not distinguish between whole grain wheat and refined wheat.
METHODS: Population-specific data from 170 countries on prevalence of obesity, availabilities of mixed cereals, wheat, rice, maize, meat, sugar, fat, soy and calories and GDP are obtained from the UN agencies. All variables were measured as per capita per day (or per year). Each country is treated as an individual subject. SPSS v. 22 is used to analyse these data for all the 170 countries and official country groupings (regions) using non parametric and parametric correlations, including partial correlation analysis.
RESULTS: Pearson's correlation coefficient analysis showed that obesity prevalence is positively associated with wheat availability (= 0.500,< 0.001), but is inversely associated with availabilities of total cereals (= -0.132,= 0.087), rice (= -0.405,< 0.001) and maize (= -0.227,= 0.004). These associations remain in partial correlation model when we keep availabilities of meat, fat, sugar, soy, caloric intake and GDP statistically constant. Overall, positive associations between wheat availability and obesity prevalence remain in different regions. Maize and mixed cereal availabilities do not show independent associations with the obesity prevalence.
CONCLUSIONS: Our study suggests that wheat availability is an independent predictor of the obesity prevalence both worldwide and with special regard to the regions of Africa, Americas and Asia. Future studies should distinguish between possible influence of whole grain and ultra-processed refined wheat products.
You, Wenpeng