Diabetes Distress and Glycemic Control: The Buffering Effect of Autonomy Support From Important Family Members and Friends

Abstract

Objective: To examine whether autonomy support (defined as social support for an individual's personal agency) for diabetes management from informal health supporters (family/friends) reduces the detrimental effects of diabetes distress on glycemic control.

Research design and methods: Three hundred eight veterans with type 2 diabetes and one or more risk factors for diabetes complications completed a survey that included measures of diabetes distress and perceived autonomy support from their main informal health supporter. Hemoglobin A_1c (HbA_1c) data from 12 months before and after the survey were extracted from electronic medical records. Linear mixed modeling examined the main effects and interaction of autonomy support and diabetes distress on repeated measures of HbA_1c over the 12 months after the survey, controlling for mean prior 12-month HbA_1c, time, insulin use, age, and race/ethnicity.

Results: Diabetes distress (B = 0.12 [SE 0.05]; P = 0.023) was associated with higher and autonomy support (B = -0.16 [SE 0.07]; P = 0.032) with lower subsequent HbA_1c levels. Autonomy support moderated the relationship between diabetes distress and HbA_1c (B = -0.13 [SE 0.06]; P = 0.027). Greater diabetes distress was associated with higher HbA_1c at low (B = 0.21 [SE 07]; P = 0.002) but not high (B = 0.01 [SE 0.07]; P = 0.890) levels of autonomy support.

Conclusions: Autonomy support from main health supporters may contribute to better glycemic control by ameliorating the effects of diabetes distress. Interventions that reduce diabetes distress and enhance the autonomy supportiveness of informal supporters may be effective approaches to improving glycemic control.

Figure 1

Simple slopes depicting the relationship of diabetes distress to prospective 12-month HbA_1c at high and low levels of autonomy support (A) and region-of-significance plot representing the same interaction (B) while controlling for age, race/ethnicity, insulin use, prior average 12-month HbA_1c, and support satisfaction. The dashed line in B represents the boundary value of the mean-centered autonomy support variable below which the simple slope between diabetes distress and 12-month HbA_1c is significant (P < 0.05).