Clinical features and drug induced side effects in early versus late antidepressant responders.

Early antidepressant response (2nd week) has been reported as the result of a true antidepressant effect and a predictor of subsequent stable response. With the purpose to study the clinical profile of early response/remission (2nd week) compared to late response/remission (4th-6th weeks), two independent major depressive disorder (MDD) samples (the Sequenced Treatment Alternatives to Relieve Depression or STAR*D n=1922 and an Italian sample n=171) were investigated. Patients were treated with citalopram in the STAR*D while in a naturalistic setting in the Italian sample. Depressive symptomatology was assessed by the Hamilton Depressive Rating Scale weekly in the Italian sample and biweekly by the Quick Inventory of Depressive Symptomatology Clinician Rated in the STAR*D. Logistic regression was used to investigate possible predictors of early response and the Bonferroni correction was applied. In the STAR*D, higher levels of baseline core depressive symptoms (Bech subscale) were associated with early response (p=0.00017), as well as lower baseline insomnia (p=0.003) and higher work and social functioning (p=0.001). In the Italian sample none of these variables were associated with the phenotype, but a non significant trend of lower baseline quality of life (p=0.078) was observed in late remitters. In the STAR*D late responders reported higher levels of antidepressant induced side effects, especially difficulty in sleeping (p=5.68e-13), with a non significant trend in the same direction in the Italian sample (p=0.09). The identification of late versus early antidepressant responders at the beginning of the treatment may be useful to guide therapeutic choices in clinical settings.