An exemplary study of treatment efficacy is the Collaborative Depression Study conducted by the National Institute of Mental Health (NIMH) in 1989. This study was the combined effort of investigators who were equally interested in two psychotherapy modalities as well as the pharmacological treatment of patients with depression. This approach clearly marked a change in research strategies from examining therapy as usually practiced (effectiveness) to studying "ideal" therapy as guided by manuals and competency ratings (efficacy).
In 1989, Irene Elkin and other principle investigators reported on the comparative outcomes of this study, which compared a standard reference treatment (antide-pressant imipramine plus clinical management) with two psychotherapies (cognitive-behavioral therapy and interpersonal psychotherapy). These three treatments were contrasted with a drug placebo plus clinical management control group. Although each of these two psy-chotherapies had been shown to be specifically effective with depression, this study was the first head-to-head comparison of these two psychotherapies. Since the initial publication, a large number of papers have been published regarding various aspects of this study.
Participants in this study were referred from psychiatric outpatient services, self-referrals, and other mental health facilities. An initial 560 patients were interviewed by clinical evaluators using the Schedule for
Affective Disorders and Schizophrenia interview to eliminate participants with comorbid diagnoses. Inclusion criteria included the diagnoses of a current episode of major depressive disorder, and a score of 14 or higher on an amended version of the 17-item Hamilton Rating Scale for Depression. Remaining candidates for the trial were excluded if they had additional psychiatric disorders, two or more schizotypal features, history of schizophrenia, organic brain syndrome, mental retardation, concurrent treatment, presence of specific physical illness or other medical contraindications for the use of imipramine, and presence of a clinical state inconsistent with participating in the research protocol (e.g., high suicidality). The 250 patients who passed the clinical and medical screening gave consent to be entered into the study and were randomly assigned to a treatment condition based on a separate computer-generated random order for each of three sites: Pittsburgh, Oklahoma City, and Washington, D.C.
The 28 therapists (18 psychiatrists and 10 psychologists) were carefully selected, trained, and monitored in the specific treatment they offered. A different group of therapists conducted treatment in each of the treatment and control conditions, with the exception of the two pharmacotherapy conditions (imipramine plus clinical management and pill placebo plus clinical management), which were conducted double blind by the same therapist. Patients were assigned to therapists within treatments according to availability, and each therapist saw between 1 and 11 patients.
The treatments were carefully defined and followed manuals that spelled out theoretical issues, general strategies, major techniques, and methods of managing typical problems. Clinical management was a component of both pharmacological conditions to ensure clinical care and to maximize compliance. Because the clinical management offered patients support, encouragement, and occasionally direct advice, this condition may have approximated a "minimal supportive therapy" condition. However, because specific psychotherapeutic interventions were not used, improvement attributed to the "minimal supportive therapy" is indicative of common factors present in all conditions, and not specific therapeutic interventions.
During the training phase of the study, all therapists received additional training in their respective approaches and met competency criteria in carrying out their assigned treatment. Throughout treatment, therapists were monitored to ensure their adherence to their respective approaches and to determine whether the treatments could be differentiated from one another. The treatment approaches were observed by researchers and were correctly classified more than 95% of the time, ensuring that differences found between treatments were indeed due to the difference in therapeutic approaches.
Psychotherapy sessions were each 40 to 50 min long, whereas the initial pharmacotherapy session was 45 to 60 min long and clinical management sessions 20 to 30 min long. Treatments were intended to last 16 to 20 weeks, however the total sample averaged 13 sessions. Those who completed therapy averaged 16.2 sessions, while early terminators averaged 6.2 sessions.
Eleven of the 250 patients dropped out of the study before the first treatment session. Of the remaining 239 patients entering treatment, 77 (32%) terminated before completion (had less than 12 sessions of treatment) either because they chose to or because they were withdrawn by the study staff. On investigation of differences between early terminators and those who completed treatment, it was determined that early terminators across all the treatments were significantly more severely depressed at intake than those who completed treatment.
Outcome was assessed on termination of treatment and follow-up intervals on measurements of depressive symptoms, overall symptoms, and general symptoms from both the perspective of a clinical evaluator and the client. Outcome was analyzed for pretreatment-post-treatment differences and predetermined levels of clinical recovery. Post hoc explanatory analyses were carried out comparing patients with different levels of depression severity.
Numerous comparisons were made, and the results of the study are very complex. Confounding variables influencing outcome included initial severity of the patient's condition, research site differences, and the attrition rate of the patients. However, each of the treatment conditions, including the control condition, evidenced significant improvement from pretreatment to posttreatment. The results for cognitive-behavioral therapy (CBT) and interpersonal therapy (IPT) fell in between the imipramine plus clinical management group (IM-CM), which was the most effective, and the placebo plus clinical treatment group (PLA-CM), which was the least effective, although the psychotherapy outcomes were closer to the more effective IM-CM condition.
The initial severity of depression proved to be a significant variable in treatment outcome. For the patients with less severe depression, there was no evidence of the effectiveness of any treatment over PLA-CM group. However, the patients who were more severely depressed and functionally impaired did extremely well with the IM-CM condition and poorly on the PLA-CM condition, with the psychotherapies falling in between.
The results suggested that the psychotherapies needed to be offered for more than 14 sessions with the patients with more disturbances, as many continued to show symptoms of depression at the end of treatment.
Among the more interesting findings were comparisons of the two psychotherapies with the PLA-CM control group. This latter condition was intended to control for the effects of regular contact with an experienced and supportive therapist, the general support of the research setting, and the effects of receiving a "drug" that was thought to be helpful, thus answering the question of whether the psychotherapies had any effects beyond what could be achieved through this rather extensive control condition.
There was limited evidence of the specific effectiveness of IPT and no evidence for the specific effectiveness of CBT. Surprisingly, there was also little evidence for superiority of either therapy in contrast to the placebo plus clinical management. The therapies were effective, but the patients who received the placebo plus clinical management also improved. These results are consistent with the common finding that the relationship that develops between therapist and patient is a potent force in enhancing treatment outcome. This positive alliance was found in the clinical management conditions (drug/placebo) as well as in the psychotherapy conditions.
In head-to-head comparisons of IPT and CBT, no significant differences were found in any of the major analyses or in comparison with patients with more and less severe disturbances. This similarity held up even on measures that were thought to be differentially sensitive to the two therapies. However, in some post hoc comparisons, modest evidence of specific effects could be found. Although all patient groups improved by the end of treatment, superior recovery rates were found for both IPT and IM-CM, when compared to the PLA-CM control group.
Was this article helpful?