EXISTING PRACTICE COMPARISON CONDITIONS
As Freedland et al. (2011) have pointed out, there are several problems with the traditional control group hierarchy. First, it blurs the distinction between the control and comparison functions of control groups (e.g., controlling for threats to internal validity vs. actively comparing one treatment to another treatment to see which one is superior). Further, the traditional hierarchy is based on the questionable assumption that behavioral trials can be isolated from the health care settings in which they are conducted. Because participants in behavioral intervention trials frequently have access to nonstudy health care services, this rarely holds true in any RCT targeting physical or mental health problems. As discussed in Chapter 1 of this book, interventions are conducted within complex contexts that affect study designs and treatment outcomes and that need to be taken into account (see Chapter 1, Figure 1.2).
The increased usage of RCTs within complex, real-world environments has necessitated the development of other types of comparison conditions. One of these is known as an existing practice comparison condition. These are used to compare experimental interventions to existing treatments or clinical practices. Existing practice controls have played an important role in medical trials, but not in behavioral trials until recently. They are referred to by many different names, including treatment-as-usual (TAU), usual care (UC), standard care (SC), or routine care (RC) (Freedland et al., 2011; Thompson & Schoenfeld, 2007).
TAU controls use the treatments that are already in place by clinicians in the settings from which participants are recruited (Mohr et al., 2009). This implies that most people with the target problem in those settings ordinarily receive a particular treatment, which may not always be the case. The TAU approach to control is most often used to label control groups in psychotherapy studies, mental health services research, and behavioral intervention trials for substance abuse (Freedland et al., 2011). UC control groups are roughly equivalent to TAU controls. In this case, their use does not imply that participants receive a specific treatment of the target problem, although that may be true in some cases. For example, the UC for an RCT on a treatment for depression may encompass a wide variety of antidepressant medications, cognitive behavioral therapies, and clinical monitoring. Usually, the UC is determined by health care providers who are independent of the research team, but not always. A UC control may also be enhanced or modified by the researcher, that is, enhanced usual care (EUC) or constrained usual care (CUC) to overcome methodological or ethical problems that would be associated with ordinary UC (Freedland et al., 2011).