Letter

In Reply: We thank Dr. Luchins for his thoughtful response to our article. We are grateful too for his injunction to consider the broader psychotherapy literature, which is often ignored in research on assertive community treatment and other treatments for serious mental illness. His principal contention—that nonspecific elements of assertive community treatment are implicated in the clinical success of a team—would appear inarguable. Nevertheless, Dr. Luchins also seems to suggest that fidelity is an unparsimonious irrelevancy. Is it necessary to take an either-or approach—to consider that favorable consumer outcomes are attributable either to nonspecific elements of assertive community treatment or to high-fidelity implementation? We would argue that each likely contributes—independently—to outcomes.

However, published research on assertive community treatment's effectiveness has not definitively determined which specific (and nonspecific) clinical and organizational elements predict better outcomes for consumers. We suspect that a richly funded, smartly regulated mental health system staffed with empowered and enthusiastic leaders who oversee clinically competent staff would likely lead to positive consumer results. The unanswered empirical question is not whether but to what extent nonspecific and fidelity-based factors independently contribute to positive outcomes.

The critical role of fidelity measures is both intuitive and empirically established: we cannot determine a program's effectiveness if we are unsure that the program has been implemented as intended. However, fidelity measures can serve many functions—both in research and practice, each of which reveals its own set of limitations. Because the Dartmouth Assertive Community Treatment Scale ( 1 ) was initially developed for research purposes and aimed to differentiate programs that were implementing assertive community treatment from those that were not, there was a greater emphasis on including features deemed both critical and unique to the model. As a result, many important clinical features relevant to any treatment model, such as clinically competent staff, were not included in the measure. Such exclusions served the aim of creating a program yardstick but undermined an alternative function of fidelity measures as a comprehensive program assessment.

A recently developed fidelity measure, the Tool for the Measurement of Assertive Community Treatment ( 2 ), addresses some of the limitations of previous tools by comprehensively assessing both organizational structures and clinical processes that experts agree reflect high-fidelity assertive community treatment.

As states move toward broader implementation of assertive community treatment and adoption of published fidelity criteria as service standards, we caution administrators to consider the overall lack of evidence in regard to which particular elements are most critical to good consumer outcomes and in regard to the optimal dose of each element (for example, whether a team with a 100-consumer caseload requires two or four nursing staff). Blind adoption of high-fidelity standards may be unnecessarily expensive.

Although now considered an evidence-based practice, assertive community treatment requires additional investigation, especially in light of an evolving, and in many regions deteriorating, service system.