Profiles of disordered eating behaviour in type 1 diabetes using the DEPS-R and behaviour and glycaemic outcomes in a real-life setting.

Abstract

The Diabetes Eating Problems Survey - Revised (DEPS-R) is commonly used to assess disordered eating behaviour (DEB) in individuals with type 1 diabetes and has advantages compared to other measures not specifically tailored to diabetes. A score ≥20 on the DEPS-R is used to indicate clinically significant DEB; however, it does not distinguish between eating disorder (ED) phenotypes necessary to guide treatment decisions, limiting clinical utility. The current study used latent class analysis to identify distinct person-centred profiles of DEB in adults with type 1 diabetes using the DEPS-R. Analysis of Variance with Games Howell post-hoc comparisons was then conducted to examine the correspondence between the profiles and binge eating, insulin restriction and glycaemic control (HbA1c, mean blood glucose, and percent time spent in hyperglycaemia) during 3 days of assessment in a real-life setting. Latent class analysis indicated a 4-class solution, with patterns of item endorsement suggesting the following profiles: Bulimia, Binge Eating, Overeating and Low Pathology. Differences in binge eating, insulin restriction and glycaemic control were observed between profiles during 3 days of at-home assessment. The Bulimia profile was associated with highest HbA1c and 3-day mean blood glucose. There are common patterns of responses on the DEPS-R that appear to reflect different ED phenotypes. Profiles based on the DEPS-R corresponded with behaviour in the real-life setting as expected and were associated with different glycaemic outcomes. Results may have implications for the use of the DEPS-R in research and clinical settings.