Martin M. Katz: Depression and Drugs

Martin M. Katz’s reply to Donald F. Klein’s final comment


Don Klein’s final comment on my book confines itself to a very small part of the case presented that early reactivity to antidepressant drug treatment, i.e., within two weeks, predicts outcome response in clinical trials. My case on this issue included reference to our Texas study which, although utilizing a moderately sized sample of drug-treated patients (n=50), demonstrated how early actions on specific behavioral components of the disorder could predict outcome (the study was designed to test hypotheses about drug actions), but the study within its limitations could not as we previously, pointed out, by itself, make the case. We cited the study evidence looking toward testing the prediction hypothesis in a “prospective” study that would include a significantly larger, and presumably, more representative sample of these diverse disorders. For this preliminary study, we enlarged the patient sample by combining the paroxetine and desipramine treated-subsamples as “antidepressants” to conduct this pilot trial of the predictability of the components. The best predictive model for the DMI treated patients had only the dimension of depressed mood-retardation after one week as the independent variable. It achieved a combination of sensitivity and specificity of 0.90 and 0.88, respectively (Katz et al 2004). The Texas study was, therefore, not intended to be a definitive study for shortening the trial period for antidepressants but simply to show the potential of this approach and provide the basis for the conduct of a large sample prospective study. I have no great argument with Klein on that as I and others, specifically L. Morey, have discussed and answered the criticism in earlier exchanges on the issue.


What is difficult to understand in Klein’s critique is his use of his disagreement with the Texas study analysis to downplay the overwhelming evidence on this issue, the results of a large body of work, conducted over decades, in order to maintain his point of view that early reactivity as a predictor has not been proven.  In the face of the evidence from other large sample studies with diverse antidepressants, proper placebo controls, and application of the established Hamilton Scale, studies such as Stassen et al (1997) and Szegedi et al (2009), in which 90% of patients who do not show any improvement after two weeks of treatment are almost certain to not show improvement or clinical response at outcome. Other studies show 70% of the drug effect in treatment-responsive patients to occur in the first two weeks and that 70% of those who do show improvement at two weeks respond clinically at outcome. Certainly these studies conducted during the past three decades, provide in sum, clear results on this issue. They confirm that early reactivity in the form of early improvement by the end of two weeks, is highly associated with type of clinical outcome.


These studies are reviewed in some detail in my book and can be examined directly in the references provided. Either Don Klein has not actually read the Stassen et al,  Szegedi et al, and Taylor et al (2004)  papers or for other reasons, refused to accept the results or is simply in denial of the factual evidence here.  It is somewhat of a mystery and may be to others who have followed this dialogue, that such an astute and experienced clinical investigator can decline to accept the established data here. I encourage the reader to examine the sources directly, in order to assess the validity of the “predictive” hypothesis.

It is important if we are to move ahead in research on clinical trials to consider alternative approaches, such as those elaborated in my recent book, Clinical Trials of Antidepressants, a follow-up to Depression and Drugs. The book is designed to apply the advances in understanding the disorder and the bases of drug action to the practice of clinical trials.


I thank the participants in this dialogue and Dr. Klein for their close attention to the issues and to the book.



Katz MM, Tekell J, Bowden CL Brannan S, Houston JP, Berman N, Frazer A. Onset and early behavioral effects of pharmacologically different antidepressants and placebo in depression. Neuropsychopharmacology 2004; 29: 566-79.

Katz MM, Berman N, Bowden CL, Frazer A. The componential approach enhances the effectiveness of 2-week trials of new antidepressants. J Clin Psychopharmacology 2011; 37:  193-218.

Stassen HH, Angst J, Delini-Stula A. Delayed onset of action of antidepressant drugs? Survey of recent results. Eur Psychiatry 1997: 12: 166-76.

Szegedi A, Jansen WT, van Wugenburg AP. Early improvement in the first two weeks as predictors of treatment outcome in patients with major depressive disorder: a meta-analysis including 6,562 patients. J Clin Psychiatry 2009; 70: 344-53.

Taylor MJ, Freemantle N, Geddes JR, Bhagwagar Z Early onset of SSRI antidepressants: Systematic review and meta-analysis. Arch Gen Psychiatry 2006; 63: 1217-23



Martin M. Katz

October 20, 2016