Multilevel approaches to increase fruit and vegetable intake in low-income housing communities: final results of the ‘Live Well, Viva Bien’ cluster-randomized trial

Abstract Background Fruit and vegetable (F&V) intake can reduce risks for chronic disease, but is much lower than recommended amounts in most Western populations, especially for those with low income levels. Rigorous research is needed on practical, cost-effective interventions that address environmental as well as personal determinants of F&V intake. This paper presents the results of a cluster randomized controlled trial evaluating the efficacy of ‘Live Well, Viva Bien’ (LWVB), a multicomponent intervention that included discount, mobile fresh F&V markets in conjunction with nutrition education. Methods Fifteen subsidized housing sites in Providence County, Rhode Island (8 intervention and 7 control sites) were randomized using a random number generator. Of these, nine housed elderly and/or disabled residents and six housed families. A total of 1597 adult housing site residents (treatment n = 837; control n = 760) were enrolled (73% women, 54% Hispanic, 17% black, Mean age 54 years). A year-long multicomponent intervention including mobile F&V markets plus nutrition education (e.g. campaigns, DVDs, newsletters, recipes, and chef demonstrations) was implemented at intervention sites. Physical activity and stress interventions were implemented at control sites. Follow-up occurred at 6 and 12 months. The main outcome measure was F&V consumption measured by National Cancer Institute’s ‘Eating at America’s Table All Day Screener’. Results From baseline to 12 months, the intervention group increased total F&V intake by 0.44 cups with the control group decreasing intake by 0.08 cups (p < .02). Results also showed an increased frequency of F&V eating behaviors compared to the control group (p < .01). There was a clear dose response effect of the F&V markets with participants who reported attending all or most of the markets increasing F&V intake by 2.1 cups and 0.86 cups, respectively compared with less than half cup increases for lower levels of market attendance (p < .05). Use of the DVDs, recipes and taste-testings was also associated with greater increases in F&V intake; however, use of other educational components was not. Conclusions LWVB is the first cluster, randomized controlled trial to demonstrate the efficacy of year-round F&V markets on improving F&V intake for low-income adults, which provides an evidence-base to bolster the mission of mobile produce markets. Further, the results more broadly support investment in environmental changes to alleviate disparities in F&V consumption and diet-related health inequities. Trial registration number Clinicatrials.gov registration number: NCT02669472