The article by Kaneriya and colleagues1 in this issue of JAMA Psychiatry reports on a randomized double-blind clinical trial examining the moderating factors that influence remission with aripiprazole treatment in an elderly population of participants with treatment-resistant late-life depression. The participants initially received extended-release venlafaxine hydrochloride to establish treatment resistance, and then they were randomly assigned to 12 weeks of augmentation with aripiprazole or placebo.2 Pursuing their a priori hypotheses, the investigators tested for the moderating effects of executive dysfunction (specifically, measures of set shifting and response inhibition), comorbid anxiety, and medical burden. All of these factors have been implicated in contributing to poor antidepressant responses in late-life depression,3,4 but they have not previously been examined in second-line therapies and often not in the context of placebo-controlled trials. The study team found that neither medical morbidity nor response inhibition was related to remission. A higher severity of anxiety predicted an overall lower remission rate but did not specifically moderate aripiprazole efficacy. In contrast, set-shifting performance measured with the Trail Making Test moderated aripiprazole efficacy; participants who performed better were more likely to remit with aripiprazole than with placebo. For participants performing poorly, aripiprazole did not demonstrate a benefit over placebo. These findings have several implications and raise many questions.