Comment by Elemer Szabadi


Comment by Elemer Szabadi

In the debate between Martin Katz and Donald Klein, the question was raised whether the finding of an antidepressant effect during the first couple of weeks of a clinical trial of an antidepressant “could be applied to individual patients” (see Katz, INHN February 26, 2015).   Some observations made by us a few years ago (Szabadi et al., 1976) may be relevant to this question. We measured pause time (total time taken up by the nine inter-phonation pauses in a sample of automatic speech, i.e. counting from one to ten) in depressed patients both before, during and after treatment with an antidepressant (amitriptyline); speech measures (both phonation and pause times) and Hamilton Depression Scores were recorded at weekly intervals. There was a close correlation between pause time and depression scores: as the depression subsided, the counting speeded up, until, on recovery from depression, it stabilized at a level characteristic of the patient.  Although initially we proposed pause time as a measure of psychomotor retardation, later, due to its close coupling to mood, we suggested that, in fact, it may be “an objective sensitive correlate of mood” (Szabadi and Bradshaw, 1983). In our early study we already found, albeit in a small number of individual patients and without a placebo control, an appreciable reduction in both Hamilton scores and pause time, during the first week of treatment.   In fact, the scores declined monotonically until full remission was observed after about six weeks. The usefulness of speech pause time as an index of psychomotor retardation was confirmed by others (Greden and Carroll, 1980; Greden et al., 1981; Hoffmann et al., 1985; for a comprehensive review, see Bennabi et al., 2013.)  It is of interest that other authors have also found reductions in this measure early in the course of antidepressant treatment. For example, Greden and Carroll (1980) commented: “of potential clinical importance, in two cases we documented decreases in pause time of more than 50% within several days after starting treatment”.  Interestingly, a similar degree of reduction in depression and pause time scores was observed in the patient whose data we published in a figure (Szabadi et al., 1976). However, not every patient may show such dramatic improvement early on in the course of antidepressant treatment, and, although improvement may start soon after initiation of treatment, it may not be large enough to be of clinical significance.  The issue of onset of action is not unique to antidepressants.  Although the antihypertensive action of beta blockers has an instantaneous onset, in cases of severe hypertension an additional antihypertensive effect develops in the course of treatment (page 369 in: Cruikshank and Prichard, 1994).  Could the onset of the clinical action of antidepressants also be related to the severity of the condition to be treated?


Bennabi D, Vandel P, Papaxanthis C, et al. Psychomotor retardation in depression: a systematic review of diagnostic, pathophysiologic, and therapeutic implications. Biomed Res Int 2013; 158746.

Cruickshank JM, Prichard BNC. Beta-blockers in Clinical Practice. Edinburgh: Churchill Livingstone; 1994.

Greden JF. Albala AA, Smokler IA, et al.  Speech pause time: a marker of psychomotor retardation among endogeneous depressives. Biol Psychiatry 1981; 16: 851-9.

Greden JF, Carroll BJ. Decrease in speech pause times with treatment of endogeneous depression. Biol Psychiatry 1987; 15: 575-87.

Hoffmann GM, Gonze JC, Mendlewicz J.  Speech pause time as a method for the evaluationof psychomotor retardation in depressive illness. Br J Psychiatry 1985; 146: 535-8.

Szabadi E, Bradshaw CM. Speech pause time: behavioural correlate of mood. Am J Psychiatry 1983; 140: 265.

Szabadi E, Bradshaw CM, Besson JAO. Elongation of pause-time in speech:  a simple, objective measure of motor retardation in depression.  Br J Psychiatry 1976; 129: 592-7.


Elemer Szabadi