Randomised control trials (RCTs) were eligible for inclusion if they compared psychological treatments with wait list, care as usual, oral placebo or psychological placebo in adults (18 years or greater) with a depressive disorder or elevated level of depressive symptomatology. Psychological treatments were defined as interventions in which the core element was verbal communication between therapist and patient or bibliotherapy with some personal support (such as telephone or email) from a therapist. Comorbid medical or psychiatric disorders were allowed. Studies aimed at maintenance treatments and relapse prevention, studies of therapies in which the psychological intervention could not assessed independently from other elements of the intervention (such as managed care interventions and disease management programmes) and studies in which the standardised effect size could not be calculated were excluded.
Participants in more than half the studies met diagnostic criteria for a mood disorder; another definition of depression (typically a high score on a self-report questionnaire) was used in the other studies. For high-quality studies only, interventions included cognitive-behavioural therapy (with or without depression recurrence prevention programme), psychodynamic therapy, counselling, behavioural activation, problem-solving therapy, self-management therapy and interpersonal psychotherapy. Most studies involved individual rather than group therapy. Diagnoses ranged from major depressive disorder (with or without atypical features), minor depressive disorder, adjusted disorder and dysthymia. Participants included women with postpartum depression, adults in general, male veterans with chronic combat-related post-traumatic stress disorder and depressive disorder, low-income, young minority women and older adults. Beck Depression Inventory (BDI) and Hamilton Rating Scale for Depression (HAMD) were the most frequently used rating scales, others included Edinburgh Postnatal Depression Scale, Montgomery-Asberg Depression Rating Scale, Geriatric Depression Rating Scale and Hopkins Symptom Checklist Depression Scale.
The authors did not state how papers were selected for the review.