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

Disclosures

BMC Public Health. 2016;16(198) 

In This Article

Results

Participants and Response Rates

In the population with type 1 diabetes, 21 people were excluded because they were unable to be reached and 14 people had died or reported that they did not have type 1 diabetes. Of the remaining 3591 people, 2419 completed the survey, corresponding to a response rate of 67 %. Non-respondents were younger, consisted of fewer women, had higher HbA1c and longer duration of diabetes, compared with respondents. In the population with type 2 diabetes, 59 people were excluded because they had died, had dementia or reported that they did not have diabetes. One thousand eighty-four answered the questionnaire, corresponding to a response rate of 54.0 %. Non-respondents were older, consisted of more women and had higher HbA1c whereas diabetes duration did not differ significantly, compared with respondents.

Socio-demographic Characteristics (Gender, Age, Education Level and Diabetes Duration)

Men were overrepresented among people with type 2 diabetes, compared to people with type 1 diabetes (Table 1). People with type 1 diabetes were generally younger, had longer diabetes duration. People with type 2 diabetes had a lower education level than people with type 1 diabetes (Table 1).

Initial Analyses of Education Level and Gender as Possible Effect Modifiers in the Models

Initially, we tested education level and gender as possible effect modifiers in relation to all outcomes. When testing outcomes of social relations we used the method by Anderson et al. We found no strong indication for education level as an effect modifier of the association between diabetes type and social relations (RERI scores between −0.60 and 0.21 and all confidence intervals included 0), and, as a result, education level was included as a covariate in subsequent logistic regression analyses. Regarding gender, analyses suggested that gender was an effect modifier in the association between diabetes type and social relations, and the logistic regression analyses were stratified according to gender (RERI scores between −1.48 and 1.64 where most confidence intervals were above or below 0).

In generalised linear regression models, no significant interaction terms were found between diabetes type and education level with regard to diet (p = 0.7339) or physical activity (p = 0.2109), and we consequently included education level as a covariate in the analyses. The p values for the interaction term for gender and diabetes type in relation to diet and physical activity were respectively 0.0050 and 0.0017, and analyses were stratified according to gender.

Social Relations (Social Network and Social Support)

For both genders, we found no differences in regard to cohabitation status by diabetes type, also after adjustment for education level (Table 2). With respect to contacts to the network, the overall pattern was that men and women with type 2 diabetes had less contact with family and friends, compared with men and women with type 1 diabetes. Differences between people with type 1 and type 2 diabetes were more pronounced among women than men. Furthermore, people with type 2 diabetes were less certain they could count on help in case of severe illness, compared to people with type 1 diabetes. For most outcomes, estimates attenuated slightly when education level and remaining social network covariates were included in the models (Table 2, Model 2–3).

Health Behaviour (Diet and Physical Activity)

Men with type 1 diabetes were more likely to eat a healthy diet compared with men with type 2 diabetes after adjustment for age and diabetes duration, whereas for women we found no significant differences (Table 3, Step 1). In men, the association between diabetes type and diet remained significant after adjustment for education level, social network variables and social support. In women, the association remained non-significant after adjustment for covariates. Only in men, we found a significant association between education level and dietary habits and between social support and dietary habits. In women, there was an association between dietary habits and cohabitation status as well as having contact with friends.

Regarding physical activity, people with type 2 diabetes were less physically active than people with type 1 diabetes (Table 3, Step 1). Associations remained significant after adjustment for education level. Differences by diabetes type were less pronounced in men after adjustment for social network covariates but remained significant (step 3). In women, the association between diabetes type and physical activity was no longer significant after adjustment for social network variables. Social support and education level were not associated with physical activity for both men and women.

Comments

3090D553-9492-4563-8681-AD288FA52ACE

processing....