Hieronymus F, et al. – Researchers conducted an item-based, patient-level, post hoc analysis to examine the effect of baseline severity of depression on the response to treatment. Specifically, they pooled and analyzed data from all completed, acute-phase, placebo-controlled, industry-sponsored, Hamilton Depression Rating Scale (HDRS)-based trials of the SSRIs citalopram, paroxetine, or sertraline in adult major depression. The study sample consisted of 8,262 patients from 28 placebo-controlled SSRI trials. Patients with a baseline 17-item HDRS sum score ≤ 18 points were defined as having non-severe depression; those scoring ≥ 27 points were defined as having severe depression. Treatment was provided with either citalopram (n = 744), paroxetine (n = 2,981), sertraline (n = 1,202), fluoxetine (active-control group; n = 754), or placebo (n = 2,581). SSRI-induced decrease in depressed mood and other HDRS symptoms belonging to the HDRS-6 subscale did not differ between patients with non-severe and severe depression. However, following exclusion of patients with rare extreme baseline values, a positive link was noted between severity and efficacy when using HDRS-17 sum score as the effect parameter. This outcome was mainly due to the observation that patients with severe depression vs those with non-severe depression display a more pronounced response to treatment with respect to non-HDRS-6 items. Relative to HDRS-6 items, non-HDRS-6 items had a higher correlation with this outcome. Furthermore, severe vs non-severe cases had non-HDRS-6 items more severe and prevalent at baseline; hence, less room was left for improvement in these areas in patients with non-severe depression.