Eight databases, including MEDLINE, PsycINFO and DARE, were searched for studies published after 1950 up to September, 2010. Additional databases were searched for cost-effectiveness studies.
Studies had to be of adults or adolescents, receiving either a low-intensity intervention delivered by para-professionals, peer supporters or psychological well-being practitioners, or less than six hours of intervention from a qualified mental health professional. Studies had to have a control group. Studies including children and participants with bipolar disorder were excluded. The main outcomes were relapse or recurrence; social function and quality of life were also assessed. The data were combined in a narrative.
Two reviewers independently selected studies. One reviewer extracted the data into a standard form, and for low-intensity studies, assessed quality using a National Institute for Health and Care Excellence (NICE) checklist. Extracted data and quality were checked by a second reviewer. Any discrepancies were resolved together, or by a third reviewer.