Letters

In Reply: We agree with Mr. Kanter that "case management" services provided by nonclinicians and "case management" interventions provided by clinicians may have little overlap beyond their shared label. To be clear that clinicians provided standard case management services in the Connecticut study, we took care throughout the report to use the term "standard clinical case management." Clinicians in both the ACT and the standard clinical case management conditions received the same training and supervision in the provision of integrated mental health and substance abuse treatment—training that presumed a clinical background that included working with people with serious mental illnesses.

The term "case management" means hundreds of different things in different parts of the country. Some have noted the paradox that, as we have become more precise in describing the components of effective case management services—most importantly, increased provision of direct service rather than referrals—we also have blurred its definition ( 1 ). Clinical case management, such as in Connecticut, includes direct provision of services by clinicians, usually master's-level social workers. At the other end of the spectrum, brokered case management, in which a case manager (typically a paraprofessional worker) helps a person enroll in services, is at best a waste of money and at worst harmful ( 1 ). Brokered case management is both prevalent and ineffective ( 1 ). On the other hand, paraprofessionals, including consumers who are not clinicians, can be effective as long as they have close consultations with appropriately skilled clinicians ( 1 ).

We share Mr. Kanter's concern that substituting nonclinicians for appropriately skilled clinicians would leave consumers in need of clinical interventions with staff who are ill equipped to meet their needs. Data from the Connecticut study and dozens of others now indicate that the ability to deliver integrated mental health and substance abuse treatment should be part of the minimal skill set for clinicians working with individuals who have co-occurring disorders. Because of the training provided in the Connecticut study, clients of clinical case managers made gains comparable to those made by ACT clients.

Mr. Seitzer's letter offers us the opportunity to underscore that evidence-based practices (such as ACT) are not evidence-based practices for everyone. Psychosocial treatments are, by definition, influenced by what else is going on in a person's environment. If rates of hospitalization are already low, assigning people to ACT cannot be expected to push these rates even lower. Rather, ACT is an excellent service for people who are hospitalized frequently ( 1 ) or coming directly from the hospital ( 2 ). We agree with Mr. Seitzer that, in addition to monitoring processes, it is important to monitor outcomes to identify when services that were once effective are no longer producing the expected gains and for whom these evidence-based practices are most effective. For example, integrated treatment may not be suitable for individuals with co-occurring disorders who have antisocial behaviors serious enough to keep them involved with the criminal justice system ( 3 ). We would put brokered case management and day treatment programs in the "not effective" category, whereas the overall evidence indicates that ACT is still very effective for reducing hospital use in areas where hospitals remain a common treatment alternative within the service delivery system.