Predicting Medication Response in Older Depressed Patients

Duration of illness is the largest moderator of response.

What characteristics in older patients with depression are associated with good medication response? To find out, researchers analyzed individual patient-level data (N=2283; age, >60) from seven placebo-controlled medication studies of major depression examined in a 2008 meta-analysis.

Patients with moderate-to-severe depression (Hamilton Rating Scale for Depression scores, >21) and illness duration greater than 10 years had robust responses to medication (58% vs. 31% for placebo), with a medium effect size (0.54). However, this group accounted for only 385 patients (17%). Among patients with both chronic and more-severe depression, differences were even more pronounced (effect size, 0.7; number needed to treat, 4). In contrast, among patients with shorter, moderate-to-severe illnesses, medication's superiority over placebo was unimpressive (effect size, 0.09). In patients with illnesses lasting less than 2 years, medications showed no advantage over placebo. Medication–placebo differences were not associated with age, gender, or single versus recurrent episodes. Overall, more than 50% of patients responded to neither medication nor placebo.

CITATION(S):

Nelson JC et al. Moderators of outcome in late-life depression: A patient-level meta-analysis. Am J Psychiatry 2013 Apr 19; [e-pub ahead of print].