Psychosocial moderators of the impact of diabetes stigma: results from the second Diabetes MILES – Australia (MILES-2) study

Abstract

<b>Objective</b> <p>To examine the association between diabetes stigma, psychological, behavioural, and Hba1c outcomes and to investigate moderation effects of self-esteem, self-efficacy and/or social support. </p> <h3><b>Research Design and Methods</b></h3> <p>Second national Diabetes MILES–Australia (MILES-2) survey: adults with type 1 diabetes (n=959, 41% male; mean±SD age: 44±15 years), insulin-treated (n=487, 60% male; age: 61±9), and non-insulin treated type 2 diabetes (n=642, 55% male; age: 61±10). (Un)adjusted linear regression analyses tested the association between diabetes stigma (DSAS) and psychological outcomes (depressive symptoms: PHQ-8; anxiety symptoms: GAD-7; diabetes-specific distress: PAID), behavioural outcomes (healthy diet and physical activity: SDSCA), and self-reported HbA1c. Interaction effects tested whether associations varied by self-esteem (RSE), self-efficacy (CIDS) and/or diabetes-specific social support (DSS). </p> <h3><b>Results </b></h3> <p>Significant positive associations were observed between DSAS and PHQ-8, GAD-7 and PAID across diabetes type/treatment groups (all p<.001), whereby each standard deviation increase in DSAS scores was associated with an approximate half standard deviation deterioration in emotional well-being. Associations between DSAS and SDSCA and HbA1c were non-meaningful. Self-esteem moderated psychological outcomes among participants with type 1 and non-insulin treated type 2 diabetes, and diabetes distress among those with insulin-treated type 2 diabetes. Interaction effects were partially observed for social support but not for self-efficacy. </p> <h3><b>Conclusions </b></h3> <p>This study provides evidence of the association between diabetes stigma and depressive/anxiety symptoms and diabetes distress, and for the moderating effects of self-esteem and social support, among adults with type 1 and type 2 diabetes. Further research is needed to examine associations with objectively-measured behavioural and clinical outcomes. </p>