Escitalopram dose–response revisited: an alternative psychometric approach to evaluate clinical effects of escitalopram compared to citalopram and placebo in patients with major depression

Open Access
Publisher Website
Google Scholar
Abstract
In continuation of a previous psychometric analysis of dose-response data for citalopram in depression, the corresponding study data for escitalopram is of interest, since escitalopram is the active enantiomer of citalopram and because citalopram was used as the active control. Revisiting those corresponding data, the psychometric properties of the Montgomery–Åsberg Depression Scale (MADRS) and the Hamilton Depression Scale (HAMD) were investigated by focusing on the unidimensional HAMD6 and MADRS6. Effect sizes were calculated and compared for two dosages of escitalopram (10 mg and 20 mg daily) and between each of these two dosages and 40 mg citalopram daily. The results showed that the three depression scales MADRS6, MADRS10 and HAMD6 were psychometrically acceptable (coefficient of homogeneity of 0.40 or higher). In the severely depressed patients (MADRS10⩾30) a rather clear dose-response relationship for escitalopram was seen on all three scales after 6 and 8 wk of therapy. Thus, the effect size for 10 mg escitalopram ranged from 0.28 to 0.38 while the effect sizes for 20 mg escitalopram raged from 0.57 to 0.77. This difference was statistically significant (p10 (effect size 0.71 vs. 0.37). An item analysis identified ‘suicidal thoughts’ to be the most discriminating item in this respect. These results for the severely depressed patients were confirmed by the patients' self-reported quality of life evaluation. When all included patients were analysed, however, no clear dose-response relationship was seen. In conclusion, a dose-response relationship for escitalopram was seen in the severely depressed patients on all outcome scales after 6 and 8 wk of treatment. After 8 wk of treatment 20 mg escitalopram was superior to 40 mg citalopram, but not after 2 wk of treatment.

This publication has 0 references indexed in Scilit: