Using supermarket loyalty card data in a pilot randomised controlled trial of a digital intervention aimed at improving food purchasing behaviour: the Front of pack Labels Impact on Consumer Choice (FLICC) study (Preprint) [post]

Richard A Harrington, Charo Hodgkins, Monique M Raats, Gill Cowburn, Moira Dean, Aiden Doherty, Charlie E M Foster, Edmund Juszczak, Cliona Ni Mhurchu, Naomi Winstone, Richard Shepherd, Lada Timotijevic (+1 others)
2018 unpublished
BACKGROUND Most food in the UK is purchased in supermarkets and many of these purchases are routinely tracked through supermarket loyalty card data. Using such data may be an effective way to develop remote public health interventions and to measure objectively their effectiveness at changing food purchasing behaviour. OBJECTIVE The FLICC study is a pilot randomised controlled trial of a digital behaviour change intervention that incorporates tailored feedback of previous purchases of ready
more » ... s and pizzas based on traffic light labelling. The control group received web-based information on traffic light labelling. This pilot trial aimed to collect data on recruitment and retention rates and to provide estimates of effect sizes for the primary outcome (healthiness of ready meals and pizzas purchased) to inform a larger trial. METHODS Participants were recruited via email from a list of loyalty card holders held by the participating supermarket. All food and drink purchases for the participants for the six months prior to recruitment, during the six week intervention period and during a twelve week wash out period were transferred to the research team by the participating supermarket. Healthiness of ready meal and pizzas was measured using a pre-developed scale based solely on the traffic light colours on the foods. Questionnaires were completed at recruitment, end of intervention and end of wash out to estimate the effect of the intervention on variables that mediate behaviour change (e.g. belief and intention formation). RESULTS We recruited 496 participants from an initial email to 50,000 people. Only 3 people withdrew from the study and purchase data were received for all other participants. 208 participants completed all three questionnaires. There was no difference in the healthiness of purchased ready meals and pizzas between the intervention and control arms either during the intervention period (P = 0.315) or at wash-out (P = 0.594). CONCLUSIONS Whilst the FLICC study did not find evidence of an impact of the intervention on food purchasing behaviour, the unique methods used in this pilot trial are informative for future studies that plan to use supermarket loyalty card data in collaboration with supermarket partners. The experience of the trial showcases the possibilities and challenges associated with the use of loyalty card data in public health research. CLINICALTRIAL ISRCTN19316955
doi:10.2196/preprints.9910 fatcat:dzxzzmbiczcztgp5vye2aemovu