The speech analysis approach to determining onset of improvement under antidepressants.

In a study of 43 hospitalized depressive patients we investigated the course of recovery at a dense time resolution throughout the first two weeks of treatment. In addition to the assessment of the patients' psychopathology on the basis of HAMD and AMDP rating scales, speech recordings were carried out for each patient immediately before the psychiatric exploration. Our analyses yielded no evidence for a delayed onset of action of antidepressants. Onset of improvement occurred in the great majority of patients (79.1%) within the first 12 days of study, independently of the severity of depression at baseline. Early improvement was highly predictive of later outcome and could not be attributed to a few items, as score reductions were observed for virtually all HAMD items at very early stages of recovery. The analysis of the patients' speaking behavior and voice sound characteristics yielded in 62.8% of cases an essentially parallel development over time for the HAMD scores on the one hand, and acoustic variables on the other. In consequence, early improvement appears to have a biological background and is unlikely to be mainly attributable to the expectations of doctors and patients.