Energy intake in the obese - is it possible to obtain unbiased information?

Abstract

The primary aim of this thesis was to develop a valid and reproducible dietary questionnaire for large surveys with specific attention to capturing dietary intake in obese subjects. Additional aims were to compare dietary intake and eating behaviour (restrained eating, disinhibition and hunger) between obese and nonobese women, to study dietary intake and food selection patterns before and after weight reducing gastroplastic surgery and to evaluate the role of baseline leptin levels (product of the ob-gene) in relation to subsequent 4-year changes in diet and weight. Reported energy intake from the dietary questionnaire was comparable to estimated energy expenditure in both obese and nonobese subjects. Furthermore it was possible to accurately rank overweight subjects according to both energy and protein intake. Large differences in dietary intake and eating behaviour was also demonstrated between the obese and nonobese women. Obese women reported larger energy and percent fat intake and lower percent alcohol intake than nonobese women. Of the three eating behaviour factors, disinhibition was strongest independent predictor of obesity. Poor weight loss after gastroplastic surgery was not associated with a high intake of sweet foods before surgery, as has previously been suggested. However, two years after gastroplastic surgery the nutritional quality of the diet was poor, especially in those with the largest weight losses, who obtained a large proportion of their total low energy intake from sweet foods. This result underscores the need for supplementation therapy after gastroplastic surgery. Finally, high leptin levels predicted weight loss (or less weight gain) in obese women with no parental history for obesity, whereas no such association was found in the obese women with a parental history for obesity. This result highlights the complex interactions between genes and environment in the aetiology of obesity. Thus, in contrast to the obesity related under-reporting demonstrated with traditional methods, the dietary questionnaire yields dietary information from obese subjects that is at least as valid as that from normal weight subjects. The dietary questionnaire evaluated and used in this thesis has been shown to be useful for studies of diet and dietary correlates in both obese and nonobese subjects.