This paper reviews outcome studies from groups facilitated for persons living with serious mental illness. Although the research base is limited (groups posing considerable methodological challenges), available studies show that, in general, group work is as effective as individual work, if not more so. This paper describes findings from studies of short-term, long-term and self-help groups dealing with mood disorders, schizophrenia, eating disorders and borderline personality disorders over the last 10 years. The concept of ' time-effective ' group work is discussed as are the potentially harmful effects of group participation. The paper concludes by arguing that group work is an effective and less expensive treatment option that would benefit from further research to establish an evidence base, both for current practice and for the expansion of group work practice in the future.