Mediators of Cancer Communication on Fruit and Vegetable Consumption

This study investigated mediators of intervention effect in a population-based randomized trial, the North Carolina Strategies to Improve Diet, Exercise, and Screening project. We tested efficacy of two health communication strategies to promote fruit and vegetable (F&V) consumption among colon cancer survivors and a comparison group. Participants were randomized into one of four groups--control, tailored print communication (TPC), telephone motivational interviewing (TMI), or combined (TPC+TMI). 825 participants completed the baseline survey and 735 the follow-up survey (response rate 90%). Based on theories of communication and persuasion, a path model was developed to observe the relationship between perception of message relevance, message recall, trust in the message, sharing message with others, and self-efficacy on F&V consumption. The TPC+TMI intervention was directly associated with F&V consumption (β=0.68, p=0.011) and indirectly associated through perception of message relevance and recall. The TPC+TMI influenced relevance (β=0.50, p=0.016), and relevance improved recall (β=0.49, p=0.005). Greater recalls improved F&V consumption (β=0.68, p=0.014) (χ²(43, N=70)=42.96, CFI=1.00, RMSEA<.001). Message relevance also improved self-efficacy through trust. Participants who perceived more relevance had more trust (β=0.48, p=0.001) and trust improved self-efficacy (β=0.553, p=0.020). Self-efficacy increased F&V consumption (β=0.273, p=0.16). Receiving TPC+TMI was directly related to more sharing of TMI messages with others (β=0.420, p=0.001), but mediated by relevance for TPC (β=0.210, p=0.019). The TPC and TMI alone were not directly associated with F&V consumption. Determining mediators of cancer communication is important for developing more effective interventions. Further research can investigate ways to enhance message relevance, recall, and trust in interventions.