Van Veldhuizen’s article challenges the flexibility of assertive community treatment (ACT), the most influential community-based intervention for individuals with severe mental illness, and proposes a Dutch alternative, termed function assertive community treatment (FACT). Controversy is healthy in science, and serious alternative viewpoints to referred wisdom should be welcomed. FACT provides a well-articulated model, bolstered by preliminary experiences suggesting that it is not only feasible but also well received by clinicians. Moreover, ACT has been modified only minimally (e.g., by adding supported employment and integrated dual disorders interventions) over the past 30 years, even as the rest of the service system and other societal factors have changed dramatically.

Scientific disagreements should of course be resolved by data rather than by opinions, but let us highlight some of the key questions to be answered. For the moment, leave aside the possibility that FACT may be uniquely and possibly solely suited for the Netherlands inasmuch as several features of the Dutch mental health system, such as better funding and more lenient hospitalization policies, differ from the U.S. system.

The Dutch proposal identifies some key issues regarding the adaptability of ACT for community mental health care in the U.S. These include: admission and discharge criteria for ACT teams, the design of rural case management services for individuals with severe mental illness, and the nature of case management services for those who do not need ACT.

First, with regard to admission criteria, what percentage of people with severe mental illness and which ones should receive ACT services? As van Veldhuizen states, the 20% figure reported in the literature is not empirically based. The FACT experience suggesting that 80% of clients warrant ACT services at some point during a 2-year period seems high but may more realistically reflect the fluctuating needs of the population. In a U.S. study, using an administrative data base and basing estimates on hospital use, Cuddeback, Morrissey, and Meyer (2006) concluded that 51% of this population should receive ACT services. What of criteria for admission? Part of the dilemma in defining ACT admission criteria is that the reliance on prior hospitalization data is now dated. If our goal is to avert unnecessary institutionalization, then the far more common U.S. problem concerns incarceration of persons with severe mental illness (Morrissey, Meyer, & Cuddeback, in press). Preventing or reversing homelessness is also a critical concern (Salyers & Tsemberis, in press). Furthermore, with the growing recognition of the effectiveness of early interventions to prevent long-term disability (Killackey & Jackson, 2007; Nuechterlein et al., 2005), it may be time to re-conceptualize the target population for ACT completely.

Second, what about discharge criteria? From a systems perspective, Stein and Test’s (1980) original notion that ACT services should be time-unlimited appears impractical, given poorly funded services systems, and unnecessary, given long-term evidence on recovery (e.g., Drake et al., 2006). A few studies evaluating step-down ACT programs (e.g., Salyers, Masterton, Fekete, Picone, & Bond, 1998) suggest that such approaches can work. If we continue to promote ACT as an evidence-based practice, we need to develop empirical guidelines for deciding when to admit and discharge clients. The notion proposed by van Veldhuizen of team-based flexibility of intensity offers an appealing alternative.

Third, what about standards for rural case management? Nearly all observers agree that the full-fledged ACT model is impractical in rural settings. Most rural ACT programs have made significant modifications in the ACT model. Clearly we should start specifying and testing service models in such settings. Rapp and Goscha’s (2004) guidelines are a good starting point.

Finally, Van Veldhuizen raises a more fundamental question, what services should be available for individuals who most of the time do not need the intensity of ACT? He correctly identifies the conundrum that the literature has extensively examined interventions for clients who need intensive services, while it is surprisingly vague about the large majority who do not require intensive services. We need evidence-based guidelines, and we would again suggest Rapp and Goscha (2004) as a starting point. One encouraging change in the service system in the last two decades is assimilation of many once revolutionary ideas from ACT, such as meeting clients in the community, focusing on practical problems, and creating multidisciplinary treatment teams. That being said, one worry we have about adopting the FACT model is that it does not adequately specify what case management services should look like for clients who are not in crisis. If usual case management services devolve to brief contacts with clients, FACT begins to look much like the episodic treatment model in which the case management responds to crises while not fully responding to the needs of most clients.

Psychosocial interventions are almost inevitably bound by culture, economy, health systems, and other temporal factors. The robustness of ACT within the U.S. mental health system for 30 years has been remarkable, but van Veldhuizen usefully reminds us that ACT has never been comparably effective in other countries and that greater flexibility may be essential as times change. The U.S. health care system and culture have clearly changed since ACT was introduced. Many principles of ACT have been absorbed into routine services, and our beliefs regarding the values, goals, methods, and economics of community mental health have been substantially altered. As clinical researchers, we should stand with van Veldhuizen in calling for empirical examination of assumptions rather than protecting conventional clinical wisdom.