Effects of eHealth Multiple Health Behavior Intervention for Physical Activity, Healthy Diet and Weight in People with Non-communicable Diseases: A Systematic Review and Meta-analysis (Preprint)

BACKGROUND Noncommunicable diseases (NCDs) are associated with the burden of premature deaths and huge medical costs globally. There is an increasing number of studies combining a multiple health behavior change (MHBC) intervention paradigm with eHealth approaches to jointly promote weight-related health behaviors among people with NCD; yet, a comprehensive summary of these studies is lacking. OBJECTIVE This review aims to meta-analyze the effectiveness and systematically summarize the characteristics of the relevant intervention studies for improving the outcomes of physical activity, healthy diet, and weight among people with NCD. METHODS Following PRISMA guidelines, 4 electronic databases (PsycINFO, PubMed, Scopus, SPORTDiscus) were systematically searched to identify eligible articles based on a series of inclusion and exclusion criteria. Article selection, quality assessment, and data extraction were independently performed by 2 authors. The standardized mean difference (SMD) was calculated to evaluate the effectiveness of interventions for 3 intervention outcomes (physical activity, healthy diet, and weight), and subsequent subgroup analyses were performed for gender, age, intervention duration, channel, and theory. Calculations were conducted, and figures were produced in SPSS 22 and Review Manager 5.3. RESULTS Of the 664 original hits generated by the systematic searches, 15 eligible studies with moderate to high quality were included. No potential publication bias was detected using statistical analyses. Studies varied in intervention channel, intensity, and content. The meta-analysis revealed that the eHealth MHBC interventions significantly promoted physical activity (SMD 0.85, 95% CI 0.23 to 1.47, P=.008) and healthy diet (SMD 0.78, 95% CI 0.13 to 1.43, P=.02), but did not contribute to a healthy weight status (SMD -0.13, 95% CI= -0.47 to 0.20, P=.43) among people with NCDs, compared to the control conditions. Results from subgroup analysis indicated that theory-based interventions achieved greater effect than nontheory-based interventions in promoting physical activity, and interventions with traditional approaches (SMS, telephone) were more effective than those with modern internet-based approaches in promoting healthy diet. CONCLUSIONS The results of this review indicates that eHealth MHBC interventions achieve preliminary success in promoting physical activity and healthy diet behaviors among people with NCD. Future studies could improve the intervention design to achieve better intervention effectiveness. TRIAL REGISTRATION PROSPERO International Prospective Register of Systematic Reviews CRD42019118629; https://www.crd.york.ac.uk/prospero/display_record.php?RecordID=118629.