Background: Rising obesity and the associated risks of diabetes and heart disease require changes in diet to bring about healthier eating. To achieve this, people need to understand nutrition and daily requirements but are frequently confused by nutrition information on food labels. The introduction of the Food Standards Agency's ‘Traffic Light’ and the alternative ‘Guideline Daily Amount’ systems may help or further confuse the public. A previous study showed that although 63% of study participants read labels, only 25% claimed to understand them, also knowledge of nutrient requirements and functions was low (Rigby, 2004). Ransley et al., (2001) have shown that till receipts can be used to estimate fat and energy intake. The aim of the current study was to investigate whether an intervention designed to improve understanding of nutrition and labels could improve food purchases and whether these changes could be measured from till receipts.Method: Participants were recruited from the general public (80 female; 23 males) (age <30 years (23), 31–45 years (25), 46–60 years (30), >60 years (22). Subjects were randomly assigned to either the intervention group (n = 78) who were provided with an information booklet and credit card sized nutrition and labelling information to use when shopping, or the control group (n = 25), who received the information after four weeks of normal shopping. The intervention group provided an initial till receipt pre intervention and was then given the nutrition and labelling material. Further till receipts were returned from successive shopping trips over the following 4 weeks. The nine categories of food used for comparison were: fruit and vegetables, saturated fats, monounsaturated fats, polyunsaturated fats, white cereals, wholegrain cereals, processed foods, full fat and reduced fat items. Ethics approval was obtained from NWW Wales NHS Trust ethics committee.Results: Each of the nine food categories on till receipts were calculated as a percentage of the total shopping, excluding non‐food items. General linear model repeated measures analyses showed differences between study participants’ food purchases. For the intervention group, purchases in three of the nine food categories showed significant improvements: increased purchases of fruits and vegetables (P < 0.001); reduction in purchases of saturated fats (P < 0.001); and reduction of white cereal purchases (P < 0.050). The control group showed no differences in any category.Discussion: Although the intervention group did show improvements in most of the other food categories, they were not statistically significant. Positive changes were found in seven of the nine categories, with only one, wholegrain cereals, showing a decrease in purchases rather than an increase. The control group displayed a random pattern over the four till receipts, with eight categories either showing negative change or no change; only one showed a positive change. The disproportionate group sizes may mean that it is not be possible to draw firm conclusions regarding the effectiveness of the intervention.Conclusions: This study demonstrates that positive changes in improving food‐purchasing choices, as measured by till receipts, can be made by using educational interventions. Further larger studies using routinely collected supermarket data would enable the study to be replicated on a much larger scale.References Ransley, J.K., Donnelly, J.K., Khara, T.N., Botham, H., Arnott, H., Greenwood, D.C. & Cade, J.E. (1991) The use of supermarket till receipts to determine the fat and energy intake in a UK population. Public Health Nutr. 4, 1279–1286.Rigby, P. (2004) Effecting change. Understanding nutritional information. Can increased knowledge and understanding in relation to nutritional information bring about a change in eating habits? PhD Thesis WA: Bangor University.