eHealth Interventions for Fruit and Vegetable Intake: A Meta-Analysis of Effectiveness

Abstract

Background. An adequate intake of fruits and vegetables (FVI) has shown benefits for reduced risk of manifesting chronic diseases. Thus, development of effective interventions to promote FVI is paramount. Aim. To assess the effectiveness of eHealth interventions for FVI targeted to healthy population, in comparison with control groups, and identify the moderators contributing to their effectiveness. Method. A database search was conducted in April 2016 and July 2018 using MEDLINE, PsycINFO, ERIC, Google Scholar, SciELO, and RISS, yielding 19 eligible studies. Risk of bias was assessed using Cochrane's Risk of Bias Tool. Random-effects model was used to calculate effect sizes (ES) by weighted standardized mean differences. Results. This meta-analysis includes 19 studies and 6,894 participants. The most common eHealth tool used was Internet-based interventions (n = 10). Most of studies were conducted in adults (n = 11), followed by children (n = 4), and adolescents (n = 4). The minimum number of behavior change techniques (BCTs) used was one, and the maximum seven. The overall ES was small (0.26, p < .001), favoring eHealth interventions. The between-studies heterogeneity was large (I-2 = 62.77%, p < .001). Subgroup analyses showed that the components with larger ES were tailored interventions (0.27, p < .001), computer-based interventions (0.44, p < .001), and use of >= 7 BCTs (0.42, p < .001). Discussion. All studies showed a positive ES favoring interventions. Large heterogeneity could be explained in part by the number of BCTs and type of eHealth tool used. Nonetheless, more concrete evidence regarding other intervention components effectiveness was limited by small sample size. Conclusion. The use of eHealth tools for the improvement of FVI showed to be more effective compared with nonintervention and interventions not using these technologies. Nonetheless, more research is needed to determine the specific combination of intervention components that could translate into larger effectiveness.