Abstract:Purpose This meta-analysis examined relationships between social support and self-care in type 1 (T1DM) and 2 diabetes mellitus (T2DM). Methods We searched for published and unpublished studies using the following databases: PubMed, Embase, PsycINFO, the Cochrane Library, and Medline. MeSH search terms included “diabetes mellitus,” “social support,” “caregiver,” “self-care,” “self-management,” “self-care skills,” and “coping behavior.” Studies reporting correlations between social support and self-care were included. Results Initially, 2 095 studies were extracted. After eliminating duplicate and irrelevant studies, 28 studies involving 5 242 patients with diabetes were included. Of these, 22 studies examined T2DM subjects. Social support was significantly associated with self-care ( k = 28, r = .28, 95% CI: .21-.34, P < .001). Among the diabetes self-care types, the strongest effect was found for glucose monitoring ( k = 6, r = .21, 95% CI: .08-.33). The relationship between social support and self-care was stronger in T2DM ( k = 22, r = .30, 95% CI: .22-.37), relative to T1DM, samples ( k = 5, r = .22, 95% CI: .02-.38). Conclusion It was concluded that the overall effect size for social support on self-care was moderate, and its strength differed by ethnic majority within the sample, type of social support measures, and publication status.

What problem does this paper attempt to address?

This paper aims to explore the relationship between social support and diabetes self - management. Specifically, the study systematically evaluated the impact of social support on self - care behaviors of patients with type 1 diabetes (T1DM) and type 2 diabetes (T2DM) through meta - analysis methods. The main purpose of the study was to quantify the strength of the association between social support and diabetes self - management, and to assess whether this association was affected by factors such as diabetes type, sources of social support, racial/ethnic composition in the sample, self - management and social support measurement tools, publication status, and sample size estimation. ### Research Background Successful blood - glucose management is crucial for preventing diabetes complications and improving psychological well - being. However, achieving this is challenging due to the complexity of diabetes management. Diabetes self - management is a strong predictor of blood - glucose control, both for patients with type 1 and type 2 diabetes. Social support is considered one of the important factors influencing diabetes self - management, but its specific impact has not been fully quantified. ### Research Methods Researchers collected relevant studies published between January 2005 and June 2015 by searching databases such as PubMed, Embase, PsycINFO, Cochrane Library, and Medline. Inclusion criteria included that the study must report correlation data between social support and self - management, and the study subjects were patients with type 1 or type 2 diabetes. The study excluded studies that could not calculate effect sizes, qualitative studies, reviews and methodological papers, and studies that defined social support as social networks or community connections. ### Main Findings - **Overall Effect Size**: There was a moderately strong positive correlation between social support and diabetes self - management ($k = 28$, $r = 0.28$, 95% CI: 0.21 - 0.34, $P < 0.001$). - **Influence of Different Self - Management Types**: - Blood - glucose monitoring: $k = 6$, $r = 0.208$, 95% CI: 0.079 - 0.330 - Complication management: $k = 2$, $r = 0.179$, 95% CI: 0.087 - 0.268 - Exercise: $k = 7$, $r = 0.152$, 95% CI: 0.062 - 0.240 - Medication adherence: $k = 5$, $r = 0.117$, 95% CI: 0.033 - 0.198 - Diet: $k = 6$, $r = 0.110$, 95% CI: 0.055 - 0.165 - Foot care: $k = 2$, $r = 0.094$, 95% CI: 0.009 - 0.177 - **Moderator Variable Analysis**: - **Diabetes Type**: The effect size of patients with type 2 diabetes ($k = 22$, $r = 0.296$, 95% CI: 0.223 - 0.366) was larger than that of patients with type 1 diabetes ($k = 5$, $r = 0.215$, 95% CI: 0.015 - 0.399), but the difference was not significant (total Q value = 1.545, $P = 0.214$). - **Source of Social Support**: The effect size of family support ($k = 3$, $r = 0.380$, 95% CI: 0.195 - 0.539) was larger than that of other support sources (such as friends, healthcare providers) ($k = 17$, $r = 0.287$, 95% CI: 0.192 - 0.376), but the difference was not significant (total Q value = 0.838, $P = 0.360$). - **Racial/Ethnic Composition**: The effect size of studies with non - white majority samples ($k = 18$, $r = 0.327$, 95% CI: 0.249 - 0.402) was larger than that of studies with white majority samples ($k = 8$,...

The Impact of Social Support on Self-care of Patients With Diabetes: What Is the Effect of Diabetes Type? Systematic Review and Meta-analysis