110 Boston Street / Salem, MA 01970 / Phone: 978.744.7905 / Fax: 978.740.9145

Evidence-based treatments in a nutshell

Evidence-based Treatments (EBTs) are “treatments that have been tested empirically and found to have beneficial effects.” 1

Most practitioners believe their practice is based on some kind of “evidence”, including their and other's clinical experience, as well as empirically-supported theory. So what's new here?

The answer is that the current EBT paradigm defines specific types of empirical evidence that are acceptable, and also defines how that evidence is to be translated into practice. These standards are in many ways more stringent than in the past.

In the EBT paradigm,

  1. The treatment is extensively specified in the form of a written manual. EBTs are usually targeted to specific symptoms or disorders, and specific types of clients. Training is often carefully specified, too, through certification programs.
  2. The preferred form of empirical evidence is a multi-site, randomized controlled trial (RTC). In this form of research, the client is randomly assigned to either the treatment being research, or to a control condition. The control condition may be a waiting list, or "treatment as usual" (TAU), or it may be another well-specified treatment method.2
  3. Since the practitioner expects to get the same results that were generated in research trials, it is important that the treatment be replicated faithfully. This is known as "fidelity". Training clinician's to implement the model properly is one important approach to Fidelity. In addition, it is common to use some kind of check on Fidelity on an ongoing basis. This can take many forms, including rating scales completed by clients, therapists, or supervisors, or having a supervisor review videotapes.
  4. Finally, many EBTs include ongoing assessment of client outcomes as an integral part of the treatment process.

EBTs in child/adolescent/family mental health (and related fields) have proliferated in the last decade, and a substantial subset of these has matured to the point of having well developed mechanisms for dissemination (training and certification).

Efficacy research usually involves “laboratory” settings with clients selected carefully for the treatment’s target disorder or symptoms. Research therapists are often researchers and their graduate students, who are steeped in the model and invested in its success. This is different from most “real-world” clinical settings where clients typically have many problems and may have more severe problems than those in research studies. It is also different from settings where therapists have heavy investment in their past training, little time for new training, and where they must be prepared to deal with a wide spectrum of problems and disorders.

The “technology transfer” or “transportability” problem of how to move EBTs from lab settings to “real world” settings is a serious challenge for practitioners and policy makers who want to use research-based treatments with real kids; it is now a subject of much discussion and is even starting to be studied empirically (see the readings on this topic).

So far, however, most research on the diffusion of EBTs into practice has been limited to descriptive surveys of provider and policymaker knowledge, attitudes, and use of EBTs, as well as anecdotal reports of specific efforts to implement EBTs. (For examples, see annual Proceedings of the conference A System of Care for Children’s Mental Health: Expanding the Research Base, hosted by the University of South Florida. Since 2004, Chapter 2 of the Proceedings has been titled “Issues in Implementing Evidence-based Practices”.)

  1. A number of similar terms are also used, usually in related fields e.g.: empirically supported treatment, evidence-based medicine, evidence-based practices, science-based intervention models.
  2. By randomly assigning clients, researchers reduce the risks of different kinds of clients getting selected into different conditions; if people with more complex or severe problems get selected into one of the conditions, it will be hard to know if different outcomes are due to different treatments, or to the composition of the groups, or some interaction between these two factors. Multi-site trials are preferred because a finding is considered more robust when it can be replicated at different times and places, and by different researchers. In general this RTC research model is similar to that used in drug trials, except psychotherapy trials are generally not "blind"; drug researchers can make placebo pills and experimental pills look the same, but it is hard to prevent the client and a therapist from knowing in which therapy condition they are participating.
Children's Friend in action.