Inclusion and exclusion criteria were developed based on the PICOS (population, intervention, comparison, outcome, and study design) framework, as shown in Table 1. Only randomized controlled trials (RCTs) and quasi-experimental (QE) studies will be included to ensure methodological rigor and allow a more reliable assessment of intervention effects. Nonrandomized and before-and-after studies will be excluded, as the absence of control groups and randomization limits the ability to draw causal inferences about intervention effectiveness.

Operational criteria for screening will be applied to differentiate social support–based interventions from general lifestyle interventions as follows:

A data extraction form will be created by PM in Microsoft Excel to capture study characteristics, intervention details, and outcome data. Data entry will be double-checked by comparing the extracted table with the original article reports. Information will be extracted from each included study, with categories provided in Multimedia Appendix 3.

If any data are unclear or missing, the corresponding author of the study will be contacted, with 1 follow-up email, to request additional information. If no response is received within 4 weeks, the specific data or outcome will be excluded from the analysis, with the reason for exclusion documented in the review. Where feasible, sensitivity analyses will be conducted to assess the potential impact of missing data on the overall findings.

Two independent reviewers (PM and DC) will assess the risk of bias of each study using the Joanna Briggs Institute (JBI) critical appraisal tools, selected according to study design. These include the revised JBI tool for RCTs and QE studies. The revised JBI tool for RCTs evaluates 5 domains: bias related to selection and allocation; administration of the intervention and exposure; assessment, detection, and measurement of outcomes; participant retention; and statistical conclusion validity. The revised JBI tool for QE studies evaluates 7 domains: temporal precedence; selection and allocation; confounding factors; administration of the intervention and exposure; assessment, detection, and measurement of outcomes; participant retention; and statistical conclusion validity [17,18]. Each item will be rated as yes, no, unclear, or not applicable. An overall risk of bias judgment will be made for each study using the grading scale adapted from a previous systematic review, where studies with 3 or fewer “no” responses were classified as low risk, those with 4 to 6 “no” responses as moderate risk, and those with more than 6 “no” responses as high risk [19]. This approach was used since the original JBI tool does not provide explicit thresholds for determining the overall risk of bias. The Grading of Recommendations Assessment, Development, and Evaluation approach will not be applied, as this review does not aim to produce certainty of evidence ratings or guideline recommendations; instead, the focus is on evaluating methodological quality and synthesizing findings across study designs. Publication bias will be examined qualitatively when fewer than 10 studies are included and, where 10 or more studies are available, assessed formally using funnel plots and Egger regression test [20].

Study characteristics, intervention details, and outcome data will be summarized in tabular form, accompanied by a narrative synthesis. Where data are qualitative or not suitable for statistical analysis, descriptive summaries will be provided. Functional dimensions of the support, including emotional, instrumental, informational, and appraisal, will be discussed qualitatively. If at least 5 studies provide comparable outcome data, a meta-analysis will be conducted to evaluate differences between intervention and control groups. This threshold was chosen to ensure stable estimation of between-study variance and to maintain adequate power for detecting heterogeneity, as recommended in previous methodological research [20,21]. For continuous outcomes, pooled results will be expressed as mean differences or standardized mean differences when measurements differ with 95% CIs, while dichotomous outcomes will be reported using relative risks with 95% CIs.

To address heterogeneity in the measurement of diet and physical activity, outcomes will be harmonized prior to meta-analysis. We will mathematically convert reported units to uniform metrics where feasible (eg, kcal to kJ, daily to weekly amounts, etc). However, incompatible outcomes will be kept as distinct domains (eg, separating total energy intake from diet quality scores). Where possible, the standardized mean difference will be used to pool effects within specific domains.

Heterogeneity will be assessed to examine the differences between studies using Cochran Q and the I² statistic, and results will be interpreted cautiously when fewer than 10 studies are available, as heterogeneity estimates and small-study effects are less reliable in small meta-analyses [22]. If substantial heterogeneity is detected (I²>50%), subgroup analyses will be conducted, or meta-regression analyses will be performed for continuous variables. A random-effects model will be applied throughout [23]. Sensitivity analyses will be carried out to assess the robustness of pooled results.

Meta-analysis and statistical analyses will be conducted using JASP (version 0.95.4).

This review is a part of a broader study aimed at developing a social support–based intervention prototype focused on improving diet and physical activity among adults living with overweight and obesity. Subsequent stages of the project will involve patient and public involvement to shape the research direction, including qualitative interviews to explore the needs of the target population and co-design workshops to develop the intervention. Therefore, it is essential to review existing literature to identify effective interventions that incorporate social support elements for behavior change.

The findings from this review will enhance understanding of current evidence on social support–based interventions for overweight and obesity and help identify key research gaps to inform future development work. With obesity rates rising globally and dietary and physical activity behaviors remaining the major drivers of excess weight, evidence on effective support strategies is needed. This review will inform the creation of scalable, adaptable interventions that can promote behavior change worldwide.

This review protocol has notable strengths. It adheres to rigorous methodological standards, including registration with PROSPERO and compliance with PRISMA guidelines, which ensures transparency and reproducibility. The comprehensive search strategy spans 6 major databases, 1 register, and 3 gray literature sources, increasing the likelihood of capturing a broad range of relevant studies. The inclusion criteria are clearly defined, focusing on structured social support interventions targeting both dietary and physical activity behaviors. The use of validated tools for risk of bias assessment, such as the revised JBI critical appraisal instruments, enhances the reliability of quality evaluations. Additionally, the dual-reviewer approach for screening and data extraction, with a third reviewer available to resolve disagreements, further strengthens the methodological rigor.

However, this review also has some limitations. First, the inclusion of only English language publications may introduce language bias and limit the comprehensiveness of the evidence base. Second, the variability in outcome measures, particularly in how diet and physical activity are assessed, may affect the consistency and interpretability of the results. Therefore, outcomes will be harmonized before meta-analysis. Finally, it is not possible to assess the intensity of the social support component of interventions (eg, percentage of contact time). However, preliminary exploration of the data found that this was not usually reported and was deemed an inappropriate indicator of intervention quality.

This protocol will systematically review social support–based interventions for diet and physical activity among adults living with overweight and obesity to identify the most effective support mechanisms and key intervention characteristics. By integrating quantitative and qualitative evidence, it will uncover research gaps and inform the development of tailored, evidence-based public health strategies to reduce the global obesity burden. As previous reviews have not fully examined how different forms of social support affect dietary and activity outcomes in this population, our work will fill this gap and guide the design of more effective, socially informed interventions.