Visualized nutrition education and dietary behavioral change: A systematic review and meta-analysis

Effectiveness of using visual approaches in health education and its influential factors were still in debate. This study aimed to asess the effects of visualized nutrition education on dietary knowledge and behavioral changes, and factors influencing them. A comprehensive search of PubMed, EMBASE, Scopus, and Cochrane Library was conducted. Eligible studies were trials assessed effects of visualized nutrition education on dietary knowledge or behavior changes, compared with non-visualized or no education group. Fourteen studies (n = 7,259) were qualitatively analyzed and 7 of them were included in the meta-analysis. We found a higher fiber intake in both short term (1.59 g/1000 kcals, 95% CI 0.90-2.27) and long term (1.36 g/1000 kcals, 95% CI 0.64-2.09). A marginal advantage was shown in short-term fruits and vegetables consumption (F&V consumption) (standardized mean difference [SMD] = 0.08, 95% CI -0.00 to 0.16). The education effects were more pronounced when education was delivered in series (SMD_{F&V consumption} = 0.09, 95% CI 0.00-0.17), avoiding loss-framing (SMD_{Fat intake} = 0.31, 95% CI 0.10-0.51) and video modeling (SMD_{F&V consumption} = 0.23, 95% CI 0.03-0.43), with short length plus cultural adaptation. Visualized nutrition education was overall promising in improving dietary behaviors. Delivering in series, short in length, with cultural adaptation were features tended to enlarge the benefits of visualized education while loss-framing and video modeling might be avoided.