Involuntary outpatient commitment is one of the most controversial issues in mental health law. The articles in this month's special section of Psychiatric Services focus on its use in New York State, where it is referred to as assisted outpatient treatment (AOT). The articles provide important information about context. In these studies and in other studies of outpatient commitment, as well as in many areas of mental health services research, it is helpful to think of context not as a limitation but as a means of explanation. Researchers often address context only in the limitations paragraph of their research reports, worrying that too much focus on the unique context will doom publication. Yet context is not something to overcome. Much as social psychology in the tradition of Kurt Lewin considers the person in the field, we need to consider a policy or intervention in its field or context.

Understanding context and including sufficiently detailed discussions of it are essential to a balanced presentation. The articles on AOT in this issue explain this context well and allow for a consideration of how positive findings are bounded by the context in which they are found. That is, some things "work" but the reasons may be related to certain elements of context, in addition to or in lieu of the effect of the intervention in question. For example, Swanson and colleagues' analysis of concerns about "queue jumping" provide context about the implementation of AOT and how it is related to service access. The meaningful role of regional differences in the use of AOT is explored by Robbins and colleagues and included in the analyses presented in all the articles in the special section. And, as the authors indicate, these findings are specifically from New York State, where passage of the AOT law was accompanied by substantial new treatment resources, which may limit the generalizability of these encouraging findings.

Given the contentious nature of this topic and the backgrounds, interests, and motivations of various stakeholders, it is critical to make this distinction about context. Stakeholders use research for various purposes, interpreting and presenting findings to their constituencies. Insisting on context as explanatory rather than limiting can help stakeholders interpret findings in productive ways. Although we need randomized studies in this area of research, we also need studies in many and varied contexts, with a detailed and frank discussion of how context is or could be related to findings.