Prevention of Recurrent Depression With Cognitive Behavioral Therapy W e read with interest the article by Fava et al 1 on the prevention of recurrent depression with cognitive behavioral therapy and wel- come the study for addressing a neglected but impor- tant area. While the long-term use of antidepressants as prophylaxis is a very important management strategy, it does have some limitations that include the reluctance of many patients to take antidepressants on a long-term basis; possible adverse effects (even the selective sero- tonin reuptake inhibitors can have important adverse ef- fects such as sexual dysfunction); the occurrence of re- lapse in some patients even with continued medication; and the possibility of residual symptoms and impair- ment in function even without an actual depressive epi- sode. In view of these factors and the demonstrated ef- ficacy of cognitive behavioral therapy in depressive disorders, its adequate evaluation for prophylaxis in depressive disorders has been long overdue and the present study is an important contribution in this direc- tion. It is not unlikely that in the next few years cogni- tive behavioral therapy and, perhaps, other modalities of psychotherapy may be shown to have an important role in the continuation and maintenance treatment of de- pressive disorders. However, we would like to point out a few con- cerns about the present study that may merit attention in more definitive subsequent studies. In the abstract, the authors state that 40 patients were randomized while ac- tually (see “Patients and Methods” section) 45 patients were randomized and 5 dropped from the analysis later. This discrepancy and the associated failure to do intent- to-treat analysis are important drawbacks. Although it is tempting to say that the analysis is intended to evalu- ate the outcome only in those who actually obtained the treatment intended, it has been shown 2-4 that this ap- proach can lead to bias. The authors fail to state how many patients were screened, how many did not meet the inclusion criteria and for what reasons, whether any patients declined to participate, etc. The summary statement that “forty-five consecutive outpatients satisfying the criteria” 1 were en- rolled could be interpreted in different ways. These is- sues bear on the generalizability of the findings and on the ability of other researchers to replicate these find- ings. Also, the reasons for some patients who are screened but not enrolled in the study are often related to the out- come of the study. The authors state that “full doses” of the medica- tions were used while their Table 1 suggests that only up to 200 mg of tricyclic antidepressants and 150 mg of sertraline hydrochloride (Zoloft) were used (instead of the recommended maximum of 300 and 200 mg, respec- tively). Perhaps the maximum tolerated doses were used, but this is not made clear. Last, it may have been difficult to maintain the blind if the same rater evaluated each patient on as many as 10 different occasions. The authors could have tested the ability of the rater to identify correctly the treatment as- signment, which should be done in all clinical trials. 4 Notwithstanding the issues, we have raised, the study is an important contribution to the literature on the topic and we look forward to more definitive studies of this important question, from Dr Fava’s group and from others. Rajnish Mago, MD Outpatient Psychiatry Clinic University of Pennsylvania 3600 Market St, Eighth Floor Philadelphia, PA 19104 Paul Crits-Christoph, PhD Director, Center for Psychotherapy Research University of Pennsylavania Philadelphia 1. Fava GA, Rafanelli C, Grandi S, Conti S, Belluardo P. Prevention of recurrent depression with cognitive behavioral therapy: preliminary findings. Arch Gen Psychiatry. 1998;55:816-820. 2. Gardner MJ, Bond J. An exploratory study of statistical assessment of papers published in the British Medical Journal. JAMA. 1990;263:1355-1357. 3. Lee YJ, Ellenberg JH, Hirtz DG, Nelson KB. Analysis of clinical trials by treat- ment actually received: is it really an option? Stat Med. 1991;10:1595-1605. 4. Schulz KF. Grimes DA. Altman DG. Hayes RJ. Blinding and exclusions after allocation in randomised controlled trials: survey of published parallel group trials in obstetrics and gynaecology. BMJ. 1996;312:742-744. In reply We welcome this opportunity to provide further information on our study. Screening for our patients took place after clini- cal remission. This carried 2 important consequences. First, patients were treated in a naturalistic way, 1 pushing the an- tidepressants dosages as high as reasonably possible ac- cording to clinical judgment. 2 Had we endorsed a fixed treat- ment strategy with a specific antidepressant, we would have obtained a more homogeneous, but less representative, sample. The second consequence of screening after remis- sion is that it allows a better evaluation of comorbidity (cer- tain types of comorbidity wane with the abatement of de- pressive symptoms). 3 Of the 58 consecutive patients who had 3 or more episodes of unipolar depression, 6 patients (10%) were excluded for unsatisfactory clinical response and 7 (12%) met other exclusion criteria (mainly presence of per- sonality disorder and/or double [ie, and/or antecedent dys- thymia] depression). Such percentages would have been higher if the patients had been screened before treatment. LETTERS TO THE EDITOR ARCH GEN PSYCHIATRY/ VOL 56, MAY 1999 479 ©1999 American Medical Association. All rights reserved. Downloaded From: http://archpsyc.jamanetwork.com/ by a University of Pennsylvania User on 09/03/2013