zations. The majority of drug treatment providers and researchers are organized into vastly different worlds, have different missions, cultures, histories, and information needs. With some exceptions, each has distinct ways of formulating, assessing, processing, and disseminating information. What both groups have in common are: (a) drug abuse and treatment are issues of primary concern, and (b) reducing drug abuse is their primary goal. Bidirectionality must be built on these two common points.
We have come to realize the relationship between drug treatment providers and drug treatment researchers is more problematic than we thought. To call for bidirectionality between treatment providers and researchers has at least four requirements. First, treatment research has to be produced for practitioners and must be useful to them. Second, practitioners must want to work with and provide information to researchers. Third, researchers must be interested in what practitioners know and want to know. And fourth, we assume that better information exchanges between practitioners and researchers will improve client outcomes. The testimony from practitioners and researchers before our overall committee challenged all of these assumptions. The exception is the current attempt of NIDA and CSAT to bridge the gap between drug treatment practitioners and researchers. This is because the context for mutuality between practice and research has yet to be achieved. The work of this committee and the necessity to bridge the gap between practice and research are made all the more timely by congressional and public criticism of the perceived ineffectiveness of drug treatment. This criticism threatens funding for practitioners and researchers alike, and provides motivation for collaboration.
In this paper, we do three things. First, we review how community has been defined in the sociological and anthropological literatures. Second, we define what is meant by "community-based" drug treatment as distinct from other treatment contexts. Third, we discuss strategies for bridging the gap between practice and research.
There is an extensive literature on community that is useful to our committee's problems with defining community-based drug treatment. The two most commonly repeated descriptors of community are: (1) social land use—those who share common residence within specific geographic boundaries; and (2) social identity—those who identity with one another regardless of shared land use. Shared land use without social identity is not sufficient to define community, while shared identity is. Social identity is