The authors report that only a small body of research focuses on this issue; however, the findings yield some conclusions about the extent and nature of the problem; strategies for improving outcomes for people with mental illnesses under community corrections supervision; and the implications of current research for policy and practice. Regarding the extent and nature of the problem, research shows that mentally ill offenders are overrepresented among probation and parole populations, and these individuals are twice as likely as those without mental illnesses to have their community supervision revoked. The best predictors of revocation for offenders with mental illnesses are similar to predictors of revocation for those without mental illnesses; however, those with mental illnesses have more of these risk factors. Research has also found that traditional community corrections agencies do not have sufficient resources to respond effectively to offenders with mental illnesses. Research has determined that a number of community-based programs have been effective in reducing recidivism for the general population under community correction supervision, but the effectiveness of these programs has not been examined for people with mental illnesses. Still, a number of general probation/parole officer strategies and techniques have shown promise in reducing the recidivism of mentally ill clients or increasing their use of services. These strategies and techniques are described, along with six evidence-based mental health treatment practices that research has shown to improve clinical outcomes for people with serious mental illnesses; however, the effectiveness of these latter practices has not been examined for people with mental illnesses who are under community corrections supervision. A research agenda is proposed to fill the identified gaps on this issue. 102 notes and a 98-item bibliography