Effects of social support interventions on social, physical and mental health outcomes: a scoping review Authors: Jessica Kersey, Sangjin Lee, Lisa Tabor Connor, Sarah Shyres, Elnaz Alimi, Noemi Rojas Serrano, Moon Young Kim, Phan Van Sang, Noelle Casem, Christian Le, Kim Lipsey, Elena Donoso Brown Published: 2026 ABSTRACT AbstractObjectivesSocial support interventions show promise for improving health, yet evidence regarding their effectiveness and optimal approaches remains limited. We aimed to examine social support intervention outcomes.DesignScoping review.Data sourcesPubMed/Medline, Elsevier EMBASE, NIBAHL and APA PsychInfo. The search was last conducted in October 2024.Eligibility criteriaIncluded studies reported the effects of interventions on the social support outcomes, mental health or physical health of adults with chronic health conditions or disabilities. Studies were excluded if interventions did not aim to build or strengthen sustainable social support or did not provide quantitative outcome data.Data extraction and synthesisWe extracted data on intervention characteristics, intervention elements and effect sizes. We synthesised effect sizes based on outcome type and intervention provider, then examined patterns in effect sizes.ResultsInterventions (n=110) were commonly delivered by peer mentors (35.1%) or mental health professionals (35.4%), primarily targeting emotional and informational support. There were small to moderate intervention effects. Skill-building interventions delivered by health professionals appeared more effective than peer-delivered interventions, but results were inconsistent across studies and there were still meaningful effects from peer mentorship.ConclusionsEffective interventions actively facilitated meaningful support by providing resources, assistance or health behaviour encouragement. Multicomponent interventions combining professional expertise with peer mentorship may optimise outcomes. Future research should further compare intervention components and formats and establish clear mechanisms linking social support improvements to health outcomes through larger-scale effectiveness trials.Keywords: Scoping Review, Public Health, Social Cohesion WHAT IS ALREADY KNOWN ON THIS TOPIC Social support is associated with better health outcomes among people with chronic conditions and disabilities, leading to increased interest in social support interventions. However, the effectiveness of these interventions and which specific components drive improvements remains unclear. Social support is associated with better health outcomes among people with chronic conditions and disabilities, leading to increased interest in social support interventions. However, the effectiveness of these interventions and which specific components drive improvements remains unclear. WHAT THIS STUDY ADDS This review revealed that social support interventions produce small to moderate effects on outcomes, but effectiveness depends on actively facilitating meaningful support rather than simply expanding social networks. Healthcare professionals and peer mentors contribute different but complementary benefits—professionals help build skills to effectively use support, while peers provide meaningful social learning opportunities, encouragement, resources and information. This review revealed that social support interventions produce small to moderate effects on outcomes, but effectiveness depends on actively facilitating meaningful support rather than simply expanding social networks. Healthcare professionals and peer mentors contribute different but complementary benefits—professionals help build skills to effectively use support, while peers provide meaningful social learning opportunities, encouragement, resources and information. HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE, OR POLICY These findings suggest that researchers need to conduct more rigorous evaluations of social support interventions that include social support outcome measures. These findings suggest that researchers need to conduct more rigorous evaluations of social support interventions that include social support outcome measures. Introduction Social support is closely associated with a broad range of health-related outcomes, including self-management, treatment adherence, physical and mental health, adjustment to disability and overall quality of life.1,9 In response, social support has been identified as an important, modifiable factor for improving health-related outcomes as a component of holistic care that addresses both medical and social needs.10 Emerging intervention research has incorporated social support into interventions to enhance health and well-being among people with chronic health conditions and disabilities. However, evidence remains limited on the mechanisms through which social support interventions exert their effects and the extent to which they effectively improve these outcomes. Our team previously examined how social support is conceptualised, operationalised and measured in health intervention research. Our findings revealed that social support interventions were multidimensional, were primarily delivered in dyadic formats (43.9%), and primarily targeted emotional support (85.1%) and informational support (54.4%). Affectionate support and positive social interaction, while highly valued by people with chronic conditions and disabilities, were addressed less frequently. The most commonly used intervention components focused on expanding social networks and developing psychosocial support skills. The development of social networks typically involved group-based or peer-based programmes that introduced new sources of support, while interventions addressing reciprocal psychosocial support skills focused on building stress management and coping strategies at the individual, dyadic or family level. While we identified a range of social support intervention components, it remains unclear which components are most effective in improving social support and its associated physical and mental health outcomes. Further research is needed to determine which of these intervention components most consistently lead to meaningful improvements. It is also important to examine whether the involvement of healthcare professionals, peer mentors or family members may affect outcomes differently. Understanding how both intervention content and the nature of support delivery influence effectiveness will be essential for optimising future social support interventions. Clarifying these elements will support the development of targeted and personalised interventions that are more responsive to the diverse characteristics of individuals and adaptable across various settings.11,13 This paper aims to describe the effect of social support interventions on perceptions of social support and associated health outcomes. Materials and methods The design and reporting of this scoping review were guided by the Preferred Reporting Items for Systematic Reviews and Meta-Analyses Extension for Scoping Reviews.14 As a scoping review, this study generated no new data and did not require Institutional Review Board approval. We searched numerous databases (PubMed/Medline, Elsevier EMBASE, NIBAHL, APA PsychInfo) using a wide range of social support search terms; the search was last conducted in October 2024. The full search strategy is provided as online supplemental material 2.1. Studies were included if they examined the effects of a social support intervention on the social support outcomes, mental health or physical health of adults with chronic health conditions or disabilities. We defined social support interventions as those that aimed to intentionally build emotional, informational, tangible or affectionate social support by actively expanding or strengthening personal sources of support. Studies were excluded if (1) the social support that was developed or strengthened was provided by a healthcare provider, organisation or paid caregiver, (2) the article did not provide quantitative data on the outcomes of interest or (3) the study examined palliative care interventions. Two reviewers independently screened all articles for eligibility, and conflicts were resolved by a third reviewer. We extracted data on each study’s country of origin, intervention setting, study design, study population (diagnostic group), sample size and sample mean age. We also extracted data on the risk of bias in the included studies, social support intervention components, supplemental intervention components, learning mechanisms and structures, outcome measures and intervention effects. Data extraction was conducted by two independent evaluators, and conflicts were resolved by a third evaluator or group discussion, if indicated. To evaluate the risk of bias of the included studies, we administered the Revised Cochrane Risk of Bias Tool for Randomised Trials15 and the Risk of Bias in Non-randomised Studies of Interventions.16 These assessments yielded an overall classification of either low risk of bias, some concerns or high risk of bias for each study. Evaluations were completed by two independent reviewers, and conflicts were resolved by a third reviewer. Our team developed a classification scheme to describe the intervention components and learning mechanisms used in each study.17 This classification scheme was generated based on our expertise in social support theory and our careful readings during the full-text review process. We pilot tested an initial set of intervention components and definitions, then made refinements based on differing interpretations and uses. We classified social support intervention components using the categories defined in table 1. Table 1Social support intervention componentsIntervention componentDefinitionDevelopment or expansion of the social networkAdding new members to the social network who can provide support (this can be through support groups or social groups, but must be intentionally developed social contacts and cannot be another person in an education group without intentional and mandatory interaction).Social activities or supported socialisationFormal or structured social activities with others or opportunities for social interaction (this can be with members of the existing social network or a new network of peers).Relationship building or strengtheningStrengthening family or friend relationships within the context of a health condition or disability (examples include family counselling or couples counselling).Communication skills and strategiesDevelopment of communication strategies to more effectively use support from the existing social network.Psychosocial support skillsDevelopment of skills or strategies for reciprocal psychosocial support, in which participants develop strategies to provide themselves and each other with psychosocial support for stress management and coping.Provision of resources or assistanceReceiving tangible resources or assistance, including help with finding and accessing resources or services.Health behaviour supportRecruitment of members of the social network to support health behaviour changes and initiation of or adherence to medical treatment plans. Social support intervention components We examined intervention effects on social support, mental health and physical health outcomes separately. We extracted or calculated Cohen’s d effect sizes using either pre-post within-group data for single-group studies and either between-groups post-intervention data or between-groups mean and distribution of change scores. Effect sizes <0.20 were considered negligible, between 0.20 and 0.50 were considered small, between 0.50 and 0.80 were considered moderate and larger than 0.80 were considered large effects.18 We analysed and narratively synthesised effects on each outcome by the type of provider (peer/layperson, health professional, self-administered, not specified) and the social support intervention components used. Results and discussion Findings A total of 110 articles were included in this review. The full results of our search and screening processes are provided as online supplemental material 2.2. Most of the included studies were randomised controlled trials (74.5%), and most interventions were delivered in health clinics or community settings (92.7%). The examined interventions were most commonly delivered by peer mentors (35.1%) and mental health providers (35.4%). Most were delivered to dyads (the patient and one supporter; 43.9%) or groups of people with a similar health condition (35.1%). Another 27.2% were delivered individually to only the person with the health condition. Full study characteristics are presented in online supplemental material 2.3. There was variability in study quality, with most studies demonstrating some concerns for risk of bias. Of the included randomised studies (figure 1), 20 had a low risk of bias (25.3%) and 4 had a high risk of bias (5.1%). The most common risk of bias was in outcome measurement. Often, this concern was due to unblinded outcome assessors. Of the included non-randomised studies (figure 2), only 3 (10.0%) had low risk of bias while 12 (40.0%) had a serious risk of bias. The most common concerns were confounding and bias in outcome measurement. The risk of bias present in the majority of the included studies suggests that findings should be interpreted with caution. Randomised studies risk of bias. Non-randomised studies risk of bias. The most commonly used social support intervention elements were the development or expansion of the social network and the development of psychosocial support skills (54.4%). The least commonly used were health behaviour change support (9.6%) and provision of resources or assistance (18.4%). Most interventions incorporated multiple social support intervention components, along with additional intervention components that did not specifically address social support. These included general health education, goal setting, psychoeducation, behaviour change strategies, chronic disease self-management education and wellness activities (eg, exercise, yoga, relaxation activities, meditation). There were also numerous structures (eg, virtual platforms) and learning mechanisms (eg, problem-solving, action planning, role playing) that supported the delivery of the interventions. These are presented in full in online supplemental material 2.4. Intervention effects for each included study are presented in online supplemental material 2.5. Online supplemental material 2.6 presents effects on each outcome of interest (social support, mental health, physical health) while illustrating the social support intervention components included in each study. Studies are further grouped by whether the intervention was delivered at least in part by peers or laypersons, or only professional healthcare providers. Overall, effects were generally small to moderate across studies, with a few interventions generating very large effects. While numerous studies were associated with moderate to large effects on outcomes, these tended to have smaller sample sizes, limiting the potential generalisability and reliability of these results. Studies with negligible effects tended to have larger sample sizes. Across all studies, peer-delivered interventions more often incorporated elements of social network development, engagement in social activities and provision of resources or assistance. On the other hand, professionally delivered interventions more commonly included elements of skill-building in relationship strengthening, communication skills and psychosocial support skills. These results suggest that peers and health professionals may enhance social support through different mechanisms, with peers serving as a source of information, encouragement and assistance, while professionals help to build skills to use social support effectively. 58 studies examined effects on social support outcomes. Of these, 17 were delivered by peers or laypersons,19,35 40 were delivered by health professionals36,75 and 1 was self-administered.76 Generally, professionally delivered interventions achieved higher effects, with 5 resulting in large effects (12.5%), 5 in moderate effects (12.5%), 15 in small effects (37.5%) and 15 in negligible effects (37.5%). Among peer-delivered interventions, 1 resulted in moderate effects (5.9%), 10 in small effects (58.9%) and 6 in negligible effects (35.3%). Peer-delivered interventions that incorporated skill-building elements and health behaviour support were associated with smaller effects on social support than those focusing more specifically on social activities or provision of resources or assistance. Effects of professional-delivered interventions that focused on skill-building elements were inconsistent across studies, ranging from very large to negligible, but these were generally the most promising interventions to improve social support outcomes. 73 studies examined intervention effects on mental health outcomes. Of these, 29 were delivered by peers or laypersons,1921,25 27 29 31 33 40 were delivered by health professionals,3639 40 42 43 47 48 51,53 55 2 were self-administered76 108 and in 2 studies, the intervention provider was not specified.109 110 Among peer-delivered interventions, 6 were associated with large effects (20.7%), 5 with moderate effects (17.2%), 6 with small effects (20.7%) and 12 with negligible effects (41.4%). Among professional-delivered interventions, 3 were associated with large effects (7.5%), 9 with moderate effects (22.5%), 16 with small effects (40.0%) and 12 with negligible effects (30.0%). Overall, peer-delivered interventions that focused on expanding the social network and providing resources and information seemed to hold similar promise at improving mental health outcomes as professional-delivered interventions addressing skill-building. 30 studies examined effects on physical health outcomes. 16 interventions were delivered by peers or laypersons,2325 29 31 77 87 88 92 111,118 12 were delivered by professionals,2137 53 56 68 102 119,124 and in 2 studies, the provider was not reported.110 125 Among peer-delivered interventions, 4 were associated with large effects (25.0%), 1 with moderate effects (6.3%), 5 with small effects (31.3%) and 6 with negligible effects (37.5%). Among professional-delivered interventions, 2 were associated with large effects (16.7%), 3 with moderate effects (25.0%), 5 with small effects (41.7%) and 2 with negligible effects (16.7%). There was a frequent focus on the development or enhancement of the social network, and this appeared promising when the social network was also specifically used to provide resources, assistance or support for adhering to health-promoting behaviours. Still, these results were inconsistent across studies, and most studies still yielded small to negligible effects. Skill-building also showed some promise, but these intervention components were examined in only a small number of studies, limiting our ability to draw conclusions about their effects on physical health outcomes. Overall, these findings suggest that social support interventions may lead to small to moderate improvements in social support outcomes, mental health and physical health. Despite inconsistent findings across studies, there are patterns suggesting that members of the social network—whether family members or peers with a similar health condition—may be able to enhance outcomes by providing resources, assistance and support for healthy behaviours. Expansion of the social network alone appears insufficient; the social network must follow through by providing informational, emotional, affectionate or tangible support to successfully improve health. These findings further suggest that health professionals can facilitate better outcomes by enhancing skills required for people with chronic health conditions to more effectively use their available social support. A combination of approaches appears most likely to improve outcomes. Our team was excited by the frequent use of peer mentors to deliver interventions. Prior research suggests that peer mentorship may enhance the perceived validity and value of interventions—peers can provide realistic advice, useful examples, meaningful encouragement and companionship based on their shared lived experience.126,129 Many of the included interventions were co-delivered by a healthcare professional and a peer mentor, which may provide the benefits of both expertise in delivering evidence-based practices, along with peer mentorship to build trust and engagement with others who have gone through similar situations. While training patients as peers may require careful thought, problem-solving and oversight, developing models to incorporate peer mentorship into evidence-based practices may be a powerful way to enhance social support and improve associated health outcomes. Our team had previously found that affectionate support and positive social interaction are critical aspects of social support that were addressed less commonly in interventions compared with social network size. This review provides some insights into how to leverage these through intervention. In the included interventions, affectionate support was often addressed through relationship strengthening, communication skills training and reciprocal psychosocial support skills training. These approaches demonstrated mixed results at improving outcomes but still appeared promising when provided by a healthcare professional. Ongoing research is needed to identify the most potent approaches to developing sustainable and effective affectionate relationships within the context of acquired health conditions and disabilities. Positive social interaction was addressed through structured opportunities for socialisation. This was typically insufficient on its own to improve outcomes and likely needs to be supplemented by additional approaches to enhance affectionate support (closeness within relationships) or the ability to use support effectively. Limitations This review had several limitations that must be considered. First, we used an inductive approach to develop our own classification scheme for identifying and describing intervention components. While we developed this classification scheme based on our expertise in social support theory and intervention research, there may be limitations in our classifications that do not capture the full breadth and nuance in social support intervention components. Second, because most interventions included multiple social support components and additional components not specific to social support, we were limited in our ability to definitively isolate the effects of specific social support intervention components. This complexity also limited our ability to identify optimal combinations of intervention components and delivery modes comprehensively. This limitation was compounded by the inconsistent intervention effects that were present even across studies with similar intervention components and structures. Together, these limitations make a comprehensive and definitive interpretation of results difficult. Our search strategy and eligibility criteria likely resulted in relevant literature being excluded from this study. We excluded studies in which healthcare providers directly provided social support, rather than working with participants to build their own external sources of support. Further research should examine whether support from providers is meaningful, sustained and improves health. We also excluded studies in which social support was used to prevent disease onset or transmission, which excluded many studies using public health and health behaviour change approaches. Further research should explore the effects of these approaches. Finally, our ongoing refinement of our definitions and terms means that we may have missed relevant studies during our screening process. Further, there are limitations within the body of included literature that must be discussed. There were concerns about the risk of bias and low study quality in many of the included studies. There was also a pattern of an inverse relationship between sample size and intervention effects. Some large effect sizes may have been found in pilot and feasibility studies and would not be replicated in larger effectiveness studies. Additionally, while social support was a critical component of the included interventions, social support was often not measured as an outcome, limiting our ability to determine whether social support improved and whether it was a meaningful piece of the mechanism leading to improvements in health. Together, these limitations affect the reliability of the findings and highlight the need for ongoing rigorous social support intervention research. It is also important to consider opportunities for the use of emerging technologies (artificial intelligence, innovative virtual settings) in the future of social support intervention research. These technologies were not well-represented in this review, but may become more relevant and widely used in future research. Finally, this study was unable to capture important cultural considerations in how social support was conceptualised and operationalised within interventions. Cultural differences in definitions, sources and uses of social support vary and must be considered in the design and implementation of appropriate social support interventions. Future research is necessary to identify optimal cultural adaptations across global regions and cultural groups. Conclusions and implications This review begins to examine the effects of social support interventions on health. Findings suggest that social support interventions may produce small to moderate improvements in social support, mental health and physical health outcomes for adults with chronic conditions and disabilities. While inconsistencies in effects and the complexity of the interventions make it difficult to draw comprehensive conclusions, there are some patterns that can inform health researchers and practitioners about how to improve social support. Findings suggest that effective interventions likely require more than simply expanding social networks—they must actively facilitate meaningful support through resources, assistance, health behaviour support or stronger support relationships. Healthcare professionals and peer mentors offer complementary strengths: professionals may effectively help develop skills for effectively using available social support, while peers provide encouragement, information, assistance and shared experiences. The most promising approach may be co-delivered interventions that combine professional expertise in evidence-based skill-building with peer mentorship to enhance social learning, appraisal, understanding and self-efficacy. Practitioners should prioritise interventions that both expand support networks and strengthen individuals’ capacity to access and use available support effectively. Critical gaps remain in understanding the active ingredients that drive intervention effectiveness. Future research must employ rigorous intervention development and reporting frameworks to clearly identify its intervention components and the specific mechanisms through which social support interventions improve health outcomes. Large-scale effectiveness trials are essential to validate promising findings from smaller pilot studies and establish the true clinical significance of social support interventions. Comparative effectiveness studies are urgently needed to compare the varied approaches to enhancing social support and determine the optimal way to improve outcomes, including the optimal intervention component combinations, delivery formats and provider types. Additionally, researchers studying social support interventions should prioritise measuring social support as an outcome variable to determine whether improvements in perceived or received social support are realised, and to establish clear causal pathways between social support intervention components and health improvements. Supplementary material Footnotes Funding: The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors. Provenance and peer review: Not commissioned; externally peer reviewed. Patient consent for publication: Not applicable. Ethics approval: Not applicable. Patient and public involvement: Patients and/or the public were not involved in the design, or conduct, or reporting, or dissemination plans of this research. Data availability statement No data are available. References Associated Data Supplementary Materials Data Availability Statement No data are available.