Digitally Delivered Interventions to Improve Nutrition Behaviors Among Resource-Poor and Ethnic Minority Groups With Type 2 Diabetes: Systematic Review

Background: Resource-poor individuals, such as those with a low income, are disproportionately affected by diabetes and unhealthy eating patterns that contribute to poor disease self-management and prognosis. Digitally delivered interventions have the potential to address some of the barriers to healthy eating experienced by this group. However, little is known about their effectiveness in disadvantaged populations. Objective: This systematic review is conducted to assess the effectiveness of digitally delivered interventions in improving nutritional behaviors and nutrition-related health outcomes among disadvantaged people with type 2 diabetes (T2D). Methods: MEDLINE complete, Global Health, Embase, CINAHL complete, Informit Health, IEEE Xplore, and Applied Science and Technology Source databases were searched for studies published between 1990 and 2022 on digitally delivered nutrition interventions for disadvantaged people with T2D. Two reviewers independently assessed the studies for eligibility and determined the study quality using the Cochrane Risk-of-Bias Assessment Tool. The Behavioral Change Technique Taxonomy V1 was used to identify behavior change techniques used in the design of interventions. Results: Of the 2434 identified records, 10 (0.4%), comprising 947 participants, met the eligibility criteria and were included in the review. A total of 2 digital platforms, web and messaging services (eg, SMS text messaging interventions or multimedia messaging service), were used to deliver interventions. Substantial improvements in dietary behaviors were reported in 5 (50%) of the 10 studies, representing improvements in healthier food choices or increases in dietary knowledge and skills or self-efficacy. Of the 10 studies, 7 (70%) examined changes in blood glucose levels, of which 4 (57%) out of 7 achieved significant decreases in hemoglobin A1C levels ranging from 0.3% to 1.8%. The most frequently identified behavior change techniques across all studies were instruction on how to perform the behavior, information about health consequences, and social support. Conclusions: This review provided some support for the efficacy of digitally delivered interventions in improving healthy eating behaviors in disadvantaged people with T2D, an essential dietary prerequisite for changes in clinical metabolic parameters. Further research is needed into how disadvantaged people with T2D may benefit more from digital approaches and to identify the specific features of effective digital interventions for supporting healthy behaviors among disadvantaged populations.