The impact of financial incentives on participants' food purchasing patterns in a supermarket-based randomized controlled trial

Background: The impacts of supermarket-based nutrition promotion interventions might be overestimated if participants shift their proportionate food purchasing away from their usual stores. This study quantified whether participants who received price discounts on fruits and vegetables (FV) in the Supermarket Healthy Eating for Life (SHELf) randomized controlled trial (RCT) shifted their FV purchasing into study supermarkets during the intervention period. Methods: Participants were 642 females randomly assigned to a 1) skill-building (n = 160), 2) price reduction (n = 161), 3) combined skill-building and price reduction (n = 160), or 4) control (n = 161) group. Participants self-reported the proportion of FV purchased in study supermarkets at baseline, 3-and 6-months post-intervention. Fisher's exact and.2 tests assessed differences among groups in the proportion of FV purchased in study supermarkets at each time point. Multinomial logistic regression assessed differences among groups in the change in proportionate FV purchasing over time. Results: Post-intervention, 49% of participants purchased >= 50% of their FV in study supermarkets. Compared to all other groups, the price reduction group was approximately twice as likely (RRR: 1.8-2.2) to have increased proportionate purchasing of FV in study supermarkets from baseline to post-intervention (p< 0.05). Conclusions: Participants who received price reductions on FV were approximately twice as likely to shift their FV purchasing from other stores into study supermarkets during the intervention period. Unless food purchasing data are available for all sources, differential changes in purchasing patterns can make it difficult to discern the true impacts of nutrition interventions.