REVIEW OF: Dunlop BW et al, Psychol Med 2019;49(11):1869–1878

TYPE OF STUDY: Post-hoc analysis of a randomized controlled trial

Antidepressants and cognitive behavior therapy (CBT) appear roughly equivalent for treating depression (Weitz ES et al, JAMA Psychiatry 2015;72(11):1102–1109), but might they differ in their efficacy for individual symptoms of depression (eg, insomnia, decreased appetite, and suicidal ideation)?

To explore this question, researchers randomized 315 depressed adults to either 12 weeks of CBT or medication treatment (escitalopram or duloxetine). They used the Montgomery-Åsberg Depression Rating Scale (MADRS) to identify residual (ie, persisting) symptoms in the 250 subjects who completed the study (n = 110 with severe depression, n = 140 with moderate depression). Subjects were categorized as responders if their MADRS score dropped by more than half during treatment. Residual symptoms from the CBT and medication groups were compared for each MADRS item, and residual symptoms of responders were compared to nonresponders.

About two-thirds of the subjects were treatment responders (59% in the CBT group and 69% in the medication group), and residual symptoms, as measured by MADRS items, were comparable among CBT and medication groups. However, among the nonresponders, CBT subjects’ mean score on suicidal ideation rose by 15%, while it dropped by 70% in the medication group—ie, even when patients failed to respond to antidepressants, their suicidal thoughts diminished.

CARLAT TAKE

Although both CBT and medications were effective for depression in this study, among treatment nonresponders, antidepressants significantly reduced suicidal ideation. CBT had the opposite effect, with suicidal ideation increasing among nonresponders.