Treatments for obesity are difficult to sustain and the response to these interventions varies between individuals. The use of direct measures of behaviour as predictive tools of weight loss may provide a personalised approach to optimise success. Our aim was to assess whether food intake from a buffet meal before and 4 months after the start of an intervention predicted weight loss.


Patients with obesity were randomised 2:1 to receive liraglutide 3mg plus a lifestyle programme (LIRA n=35) or a standard lifestyle programme (STD n=14). All patients completed a standardised buffet meal where macronutrient breakdown along with caloric intake was assessed. Participants attended at baseline, and 4, 7 and 12 months after intervention. We present our initial data collected from both groups at baseline and 4 months.


At 4 months, patients in the LIRA group lost 12.8 ±9.2kg (P=0.001) compared to patients in the STD group who lost 5.9 ± 8.1kg (P=0.06). The calorie intake from the buffet lunch meal reduced significantly in both groups at 4 months but the reduction was more prominent in the LIRA group (296kcal, p=0.002 vs 169kcal, p=0.02). Food intake at baseline was significantly associated with percentage weight loss at 4 months in the STD group (R2= 0.669, P=0.007) but not in the LIRA group (R2= 0.005, P=0.67). There were no changes in the macronutrients selection between baseline and 4 months in both groups. Weight loss at 4 months did not correlate with the type of foods consumed.


It is possible to use total food intake assessed with a buffet meal as a direct measure of behaviour. The patients who consumed the highest number of calories pre intervention lost the most weight with a standard lifestyle treatment over 4 months and thus it may be used as a predictive tool. This was however not the case for liraglutide 3mg. One year data is awaited to determine whether the change in food intake between pre-intervention and 4 months may predict longer term weight loss.