Skip to main content
. Author manuscript; available in PMC: 2017 Jul 15.
Published in final edited form as: J Affect Disord. 2016 Apr 13;199:148–156. doi: 10.1016/j.jad.2016.04.017

Appendix 2.

Prediction of Social-Interpersonal Functioning Changes from Prior Depressive Symptoms

Multilevel models: Predictors B SE p
Outcome: SAS-SR at time t0
1. Prior SAS-SR score (t-1) 0.558 0.039 <.001
 Prior depressive symptoms (t-1) −0.036 0.039 .354
Outcome: IIP at time t0
2. Prior IIP score (t-1) 0.959 0.015 <.001
 Prior depressive symptoms (t-1) −0.010 0.017 .566
Outcome: DYS at time t0
3. Prior DYS score (t-1) 0.960 0.021 <.001
 Prior depressive symptoms (t-1) −0.011 0.025 .669

Note. N = 241 patients with higher-risk response to acute-phase cognitive therapy were randomized to 8 months of continuation treatment (cognitive therapy, fluoxetine, or pill placebo with clinical management) and followed up to 24 additional months. The Social Adjustment Scale—Self-Report (SAS-SR), Inventory of Interpersonal Problems (IIP), and Dyadic Adjustment Scale (DYS) were predicted by a composite of the Hamilton Rating Scale for Depression, Beck Depression Inventory, and the Inventory for Depressive Symptomatology—Self-report at assessments taken at randomization and every 4 months through 32 months. Multilevel models controlled receipt of non-protocol treatment, continuation phase group, assessment time, and the group × time interaction.