1
Evidence-Based Practice: An Introduction

Peter Sturmey

Evidence-Based Practice

Questions concerning evidence-based practice (EBP) permeate services for people with developmental disabilities. A parent must decide whether or not to participate in early intervention for their child with autism and stay at home instead of working. A teacher selects certain target behaviors and teaching strategies for a child and decides not to teach other skills and not to use other teaching strategies. An agency for adults with disabilities decides whether to operate traditional, center-based services or to implement a new job coaching service. A city, state, or government agency decides whether or not to fund early intervention or to place some individuals in specialized, expensive, out-of-district services.

EBP is reflected in many educational and clinical decisions by individual teams. Consider the following example. A team of professionals in special education attempt to treat food refusal in a child with autism for 6 months using sensory integration therapy. Not only did the child continue to refuse food, but the child continued to lose weight. When outside therapists proposed using escape extinction (see Chapter 8) as an EBP, the educators oppose such treatment on the grounds that it does not address the sensory needs of the child and will not work or they refuse to treat the problem by “conditioning” or “behavior modification” which they think is “inhuman” or “disrespectful.” After 2 weeks of escape extinction, the child now ate a wide range of foods and gained weight; resources are no longer wasted on ineffective therapy and useless discussion of ineffective treatment.

EBP is not some academic question. If we are concerned with personally significant outcomes and avoidance of harm for individuals with developmental disabilities, it is one that we all face. This chapter outlines some of the issues in the application of EBP to services for people with intellectual disabilities (ID), autism, and other developmental disabilities. The next section examines the general and operational definitions of EBP. The next sections examine the ethical and economic rationales for EBP and the methods associated with EBP, such as systematic reviews and meta-analyses. The final section reviews some of the application of EBP to ID, autism, and other developmental disabilities.

What Is Evidence-Based Practice?

Some definitions

Some general definitions

The definition of EBP is anticipated in Paul’s (1967) famous questions: “What treatment, by whom, is most effective for this individual with that specific problem, and which set of circumstances?” (p. 111) which—nearly 50 years ago—raised the issue of not only what kinds of psychological therapy are effective but also how a practitioner should apply or not apply the results of therapy outcome research to specific clients with specific problems. Paul’s question is echoed in Sackett, Richardson, Rosenberg, and Haynes’ (1997) definition of EBP as “the integration of best research evidence with clinical expertise and patient values” (p. 1). This definition is cited very often and is the basis for similar definitions, such as those by the Institute of Medicine (2001) and the American Psychological Association’s (APA) Presidential Task Force on EBP (2006).

Table 1.1 lists a number of definitions of EBP. An examination of these definitions shows that they are aspirational rather than operational, as they do not describe the methods by which we might determine and apply best research evidence clearly. For example, the meaning of the words such as “most effective,” “integrate,” “clinical expertise,” “wisdom” and “values,” and “informed by research” is not specified in these definitions, although some sources, such as Straus, Glasziou, Richardson, and Haynes (2011), do describe specific procedures that practitioners can use to determine what is EBP and how to use it effectively with specific particular clients. Some definitions give greater emphasis to science and EBP, such as Sackett et al.’s (1997) and Dunst et al.’s definitions. Other definitions, such as Buysse and Wesley’s, appear to give equal weight to clinical experience and wisdom and only reference “best available evidence” rather than research or empirically validated treatments. Some definitions emphasize that evidence must be current and the “best available evidence” but also that this evidence must be integrated into clinical decision making. Finally, all of these definitions state that evidence and experience must be “integrated,” but these definitions do not specify what constitutes integration or how it can be achieved.

Table 1.1 Some Definitions of EBP

1. “… the integration of best research evidence with clinical expertise and patient values” (Sackett et al., 1997, p. 1)
2. “Evidence-based practice in psychology (EBPP) is the integration of the best available research with clinical expertise in the context of patient characteristics, culture, and preferences” (APA, 2006, p. 273) (http://www.apa.org/practice/resources/evidence/evidence-based-statement.pdf)
3. “… a decision-making process that integrates the best available evidence with family and professional wisdom and values” (Buysse & Wesley, 2006, p. 12)
4. EBP early childhood intervention practices are “informed by research, in which the characteristics and consequences of environmental variables are empirically established and the relationship directly informs what a practitioner can do to produce a desired outcome” (Dunst, Trivette, & Cutspec, 2002, p. 3)

These definitions illustrate the tension between the role of science and personal and professional experience should have in determining what an appropriate treatment is for a specific person. They also reflect the tension between personal and professional autonomy and choice over selection of treatment versus restriction of choice and autonomy implied by restriction of practice to only EBP by funding agencies and professional organizations. For example, the explicit aim of some meta-analyses is to determine the standards of practice in developmental disabilities (Scotti, Evans, Meyer, & Walker, 1991). Professional practice is also restricted by treatment algorithms and practice guidelines. For example, National Institute for Clinical Excellence’s (NICE) (2012) guidelines for adults with autism baldly state, “Do not provide facilitated communication” (p. 24). They also guide professionals as how to conduct certain treatments. For example, when conducting social learning programs for adults with autism, the guidelines state that they should “typically include: … modeling … peer feedback (for group-based programs) or individual feedback (for individually delivered programs) … discussion and decision-making … explicit rules … [and] … suggested strategies for dealing with difficult social situations.” A clinician who does not follow such evidence-based, professional practice guidelines would have to justify deviations from them or use of alternate approaches and might encounter censure during peer review or any legal proceedings if a client or someone else is harmed. A final important observation on these definitions is that they go beyond merely generating lists of treatment that meet the criteria for EBP. EBP also requires the application and adaption of research findings to the actual effective delivery of the best intervention to real-world clients in real-world settings by local practitioners to achieve actual client outcomes for the specific client at hand.

Operational definitions

As well as these general definitions of EBP, there are also operational definitions of EBP. Here, we can discern two kinds of approaches. The first is to determine if a specific treatment is an EBP. For example, Chambless and Hollon (1998) operationally defined an EBP in two ways. When discussing randomized controlled trails (RCTs), they write that “Only when a treatment has been found efficacious in at least two studies by independent research teams do we consider its efficacy to have been established and label it an efficacious treatment. If there is only one study supporting a treatment’s efficacy, or if all of the research has been conducted by one team, we consider the findings promising but would label such treatments as possibly efficacious, pending replication” (p. 10), and later, when discussing small N experiments, they wrote that “We consider a treatment to be possibly efficacious if it has proved beneficial to at least three participants in research by a single group. Multiple replications (at least three each) by two or more independent research groups are required before we consider a treatment’s efficacy as established (each in the absence of conflicting data).”

The second approach is to determine what an EBP is for a specific client and a specific presenting problem. Straus et al. (2011) described a five-step procedure that a practitioner should use to identify an EBP for a specific clinical situation. Step 1 was to convert an unmet information need into an answerable question. Step 2 was to find the best evidence available to answer that question. Step 3 was to appraise the evidence critically as to its validity, effect size (ES), and applicability...