One-year changes in symptoms of depression and weight in overweight/obese individuals with type 2 diabetes in the Look AHEAD study

Abstract

Depressed individuals are frequently excluded from weight loss trials because of fears that weight reduction may precipitate mood disorders, as well as concerns that depressed participants will not lose weight satisfactorily. The present study examined participants in the Look AHEAD study to determine whether moderate weight loss would be associated with incident symptoms of depression and suicidal ideation, and whether symptoms of depression at baseline would limit weight loss at 1 year. Overweight/obese adults with type 2 diabetes (n = 5,145) were randomly assigned to an Intensive Lifestyle Intervention (ILI) or a usual care group, Diabetes Support and Education (DSE). Of these, 5,129 participants completed the Beck Depression Inventory (BDI) and had their weight measured at baseline and 1 year. Potentially significant symptoms of depression were defined by a BDI score ≥10. Participants in ILI lost 8.6 ± 6.9% of initial weight at 1 year, compared to 0.7 ± 4.8% for DSE (P < 0.001, effect size = 1.33), and had a reduction of 1.4 ± 4.7 points on the BDI, compared to 0.4 ± 4.5 for DSE (P < 0.001, effect size = 0.23). At 1 year, the incidence of potentially significant symptoms of depression was significantly lower in the ILI than DSE group (6.3% vs. 9.6%) (relative risk (RR) = 0.66, 95% confidence interval (CI) = 0.5, 0.8; P < 0.001). In the ILI group, participants with and without symptoms of depression lost 7.8 ± 6.7% and 8.7 ± 6.9%, respectively, a difference not considered clinically meaningful. Intentional weight loss was not associated with the precipitation of symptoms of depression, but instead appeared to protect against this occurrence. Mild (or greater) symptoms of depression at baseline did not prevent overweight/obese individuals with type 2 diabetes from achieving significant weight loss.

Figure 1

Change in Beck Depression Inventory (BDI) score (± SEM) at 1 Year in Intensive Lifestyle Intervention (ILI) and Diabetes Support and Education (DSE) groups by depression status at baseline. D = participants who reported symptoms of depression at baseline; ND = participants with no/minimal symptoms reported at baseline. Mean (±SD) baseline BDI scores were as follows: ILI-ND 3.6 ± 2.7; ILI-D 14.4 ± 4.5; DSE-ND: 3.8 ± 2.8; DSE-D: 13.5 ± 3.7. Symptoms of depression declined significantly more (P<0.001) in the ILI than DSE group, and there was a significant (P<0.035) depression status by treatment group interaction.

Figure 2

Mean percent (± SEM) of initial weight lost at 1 Year in Intensive Lifestyle Intervention (ILI) and Diabetes Support and Education (DSE) groups by depression status at baseline. D = participants who reported mild or greater symptoms of depression at baseline; ND = participants with no/minimal symptoms reported at baseline. A 2 × 2 ANOVA (i.e., treatment-group by depression-status) revealed that ILI participants lost significantly (P<0.001) more weight than DSE participants and that participants considered free of symptoms of depression at baseline lost significantly (P<0.009) more weight than those who reported mild or greater symptoms of depression. The treatment group by depression status interaction was not significant (P=0.110).