Using a divider nudge in supermarket shopping trolleys to increase fruit and vegetable purchases: A feasibility study using an intervention design.

Abstract

Fruit and vegetable (F&V) consumption is associated with a reduced risk of developing obesity and chronic diseases: however, only one in 16 Australian adults consume F&Vs at the recommended two servings of fruit and five servings of vegetables per day. What and how much people eat is influenced by their social and physical environments. Supermarkets are a key setting influencing food purchases, and as such, they can shape consumption patterns of F&Vs. Implementing effective strategies to increase F&V intake is crucial. The objective of this research was to test the feasibility of covertly modifying shopper purchasing behaviour to purchase more F&Vs using a visual divider nudge message (prompts) covering the entire base of shopping trolleys. Placards provided a visual representation of the recommended proportion of the trolley base that should be allocated to fruits and vegetables (implied social norm). Applying an intervention research design, 30 out of ~100 trolleys were fitted with the placards and shopper purchases were measured by collecting receipts to measure the weight (kg), total spending and F&V specific spending (Australian dollars) for intervention versus control trolleys for one weekend day only. We also conducted a short intercept survey that was administered independently from the research study day on non-trial shoppers. Shoppers who selected trolleys with the divider nudge placards (n = 102) purchased equal weight of F&Vs (Intervention: mean = 6.25 kg, SD = 5.60 kg, 95% CI = 5.14 kg, 7.35 kg, vs. Control: mean 6.03 kg, SD = 5.17 kg, 95% CI = 5.01 kg, 7.04 kg, p = 0.768) and spent equal amounts on F&Vs compared to shoppers in the control group (n = 102) (Intervention: mean = $41.46, SD = $36.68, 95% CI = $34.25, $48.66, vs. Control: mean $39.85, SD = $33.30, 95% CI = $33.34, $46.39, p = 0.744). There was no difference in the total spending between groups (Intervention: mean = $135.99, SD = $90.10, 95% CI = $118.29, $153.68, vs. Control: mean $155.68, SD = $96.46, 95% CI = $136.73, $174.63, p = 0.133). The divider nudge placard did not lead to any difference in shoppers' purchases of F&Vs. However, this study demonstrates the feasibility of testing a cheap, simple and easy supermarket nutrition intervention. Larger studies are required to elucidate and confirm these findings over the longer term.