Response inhibition predicts poor antidepressant treatment response in very old depressed patients

Objective: There have been mixed findings regarding the prognostic significance of age of onset, executive dysfunction, and hyperintensity burden on treatment outcome in late-life depression. Methods: Growth curve models were fit to data from the only 8-week, double-blind, placebo controlled trial of citalopram ( 20 - 40 mg/day) in patients aged 75 years and older with unipolar depression. Baseline assessment included Structured Clinical Interview for Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition (DSM-IV) ( to determine age at onset), Stroop Color-Word Test ( to assess the response inhibition component of execution dysfunction), and structural magnetic resonance imaging ( to determine hyperintensity burden). Results: In the citalopram condition, patients with response inhibition ( most impaired quartile) scored higher at endpoint than those without response inhibition. There were no effects for age of onset or hyperintensity load on response in the citalopram condition. In the placebo condition, patients with early-onset depression had higher depression scores at endpoint than patients with late-onset depression. Conclusion: Only response inhibition, a fundamental executive function, predicted poor treatment response to antidepressant medication. Although patients with response inhibition also showed deficits in reaction time, adjusting for reaction time in our final response inhibition model did not substantively change the findings.