Randomised controlled trials (RCTs with at least 30 participants) and studies of pre-post, single-group design (with at least 80 participants), that evaluated behaviour-based interventions to promote intake of fruit and vegetables in the USA, were eligible for inclusion. Studies were only included if they explicitly reported the behaviour theory or construct used for the behaviour-based interventions. Studies of individuals with a specific clinical diagnosis (such as cancer, hypertension, diabetes or obesity) were excluded. Eligible studies had to report the outcome of fruit and vegetable intake (servings per day).
The included studies adopted a variety of behaviour theories and/or constructs applied in the interventions, such as Transtheoretical Model, social cognitive theory, behavioural construct of self-efficacy and health promotion model with tailored messages. Motivational interviewing and group/individual diet education appeared to have been the main elements of behaviour-based interventions. For controlled trials, the most common comparators were no intervention and standard programme without tailored education sessions. The dietary measurement instruments used to estimate fruit and vegetables intake also varied between studies. The most commonly used tools were food frequency questionnaires, National Cancer Institute fruit and vegetables screener and 24-hour dietary recalls. Most included studies recruited healthy adults or children; some studies recruited special populations (such as work-site-based, minority and lower-income populations).
The authors did not state how many reviewers assessed studies for inclusion.