Relationship Between Social Network, Social Support and Health Behaviour in People With Type 1 and Type 2 Diabetes

Cross-Sectional Studies

Nana F. Hempler; Lene E. Joensen; Ingrid Willaing


BMC Public Health. 2016;16(198) 

In This Article


Our findings suggest that, compared to men and women with type 1 diabetes, men and women with type 2 have less contact with family and friends and are less certain about support in case of severe illness. Differences were more pronounced among women than in men. Men with type 2 diabetes had worse dietary habits than men with type 1 diabetes, and associations were not influenced by education level, social network contacts or social support. In women, we found no differences in dietary habits according to diabetes type. Furthermore, men and women with type 2 diabetes reported lower physical activity levels than men and women with type 1 diabetes. In women, social network contacts seemed to explain differences in physical activity according to diabetes type.

To be best of our knowledge, only one study has compared social relations and health behaviour with respect to diabetes type. The study by Aalto et al. showed that men with type 2 diabetes were more likely to be single/unmarried and to report less contact with friends than men with type 1 diabetes, whereas no differences were observed for women.[11] However, in our study we found similar patterns for men and women regarding cohabitation status and contact with friends. Variations in findings across studies may be a result of differences in how social variables were measured and how analyses were performed.

Alto et al. found better dietary habits (regular meals) among people with type 1 diabetes, compared with those with type 2, particularly among men, as we did in our study. Furthermore, Aalto et al. also found that men with type 1 diabetes reported more exercise than did men with type 2 diabetes, whereas differences were less pronounced in women.[11]

The fact that people with type 2 diabetes reported fewer social relations than people with type 1 diabetes may be explained by several factors. Older age and possibly poorer lifestyle among people with type 2 diabetes may be associated with comorbidities such as cardiovascular diseases, chronic obstructive pulmonary disease and arthritis, all factors likely to increase the need for social support and making it difficult to establish, maintain and strengthen contacts with the social network. Another possible explanation for fewer social relations and poorer health behaviours among people with type 2 diabetes could suggest a reverse causation, as lack of social support and social network may promote stress and poorer health behaviour and thereby increase the likelihood of type 2 diabetes.

People with type 2 diabetes reported poorer health behaviours in terms of dietary habits and physical activity than did those with type 1 diabetes, with the exception of dietary habits in women. This was expected due to the relationship between type 2 diabetes and health behaviours. Notably, social relations only diminished the association between diabetes type and of physical activity in women slightly, underscoring the importance of addressing gender differences. In addition, it is also known from the literature that women generally have more and closer contact with a social network than do men.[24,28]

Regarding the role of education, our findings did not suggest education level as an effect modifier of the observed associations. Education level as a covariate had no strong effect with respect to social relations. However, education level seemed to play a role regarding diabetes type differences in physical activity. Regarding diet, we found that education level was highly associated with diet but did not influence the association between diabetes type and diet. This might be explained by the fact that differences in education level in our data were less pronounced in men compared with women.

Strengths of our study include that we surveyed two large populations from the same area. Survey data were merged with data from an electronic patient record which allowed comparison of respondents and non-respondents on selected variables. Limitations include the cross-sectional design which prevented us from determining causal relationships between diabetes type and the outcome variables and the influence of education level. Differences in the participation rate and characteristics of respondents and non-respondents differed by diabetes type and may cause some bias. Non-respondents in the type 1 population were younger, consisted of fewer women, and had shorter diabetes duration whereas this was reversed in the type 2 population. However, in both populations, non-respondents had higher HbA1c levels than respondents. In addition, the higher participation rate in the type 1 population might be influenced by differences in education and literacy, which are expected to be higher in the type 1 population compared to the type 2 population. Furthermore, perceived social support was measured by a single variable; according to Due et al., it is multidimensional, including social support, relational strain and social anchorage.[24] In addition, social support can also be investigated in terms of positive vs. negative, perceived available vs. perceived received and directive vs. non-directive.[29]