Eleven trials (486 participants) were included in the review: four RCTs and seven open label trials.
All four RCTs were double blind and used intention-to-treat analysis. Where reported, between 59% and 100% RCT patients completed treatment.
Two RCTs were placebo controlled; one looked for a more than 25% reduction in depressive symptoms and the second looked at the proportion of patients who met criteria for major or minor depression after treatment. Both placebo-controlled trials found no statistically significant differences between intervention and placebo groups.
One RCT compared sertraline versus venlafaxine; no statistically significant differences in response to treatment were reported as measured by the Hamilton Rating Scale for Depression and Geriatric Depression Scale. This trial showed statistically significant improvement in favour of sertraline as measured by the Cornell Scale for Depression in Dementia.
The fourth RCT was a dose response trial that looked for a more than 33% decrease in depressive symptoms; no statistically significant difference was reported between high and low dose nortriptyline.
Four of the seven open label studies reported a treatment response rate of 50% or more (range 50% to 66%); treatments included mirtazapine, sertraline, fluoxetine and paroxetine.