Steps in the Process of Evidence-Based Practice

Step 1: Framing the Clinical Question

The first step in applying evidence to a clinical decision is framing the specific question about which evidence will be sought. One widely-used approach to framing these questions is known as PICO, for Population, Intervention, Comparison, Outcome. Ensuring that the clinical question addresses all four of these areas will help to ensure that the evidence will be relevant to the particular circumstance faced by the clinician.

Population Intervention Comparison Outcome
Stroke patients Early initiation of aphasia treatment Aphasia treatment after initial/spontaneous recovery time Functional communication abilities
Kindergarteners with articulation disorders Individual pull-out treatment Group pull-out treatment Ability to consistently produce /s/
17-year-old male with a severe head injury Cognitive rehab No cognitive rehab Return to work/school

The first question, then, could be written as, "Are patients with aphasia who received SLP services shortly after their stroke more or less likely to achieve functional communication abilities than stroke patients who received such treatments later?".

The most difficult aspect of framing appropriate clinical questions is determining the level of specificity of what goes into each of these categories. To some, it is difficult because it is an art, rather than a science, and usually requires a fair amount of trial and error before the question is in its final form.


The second example was Kindergarteners with articulation disorders. Here, the clinician will need to decide whether he or she wants evidence drawn only from Kindergarteners, or would studies including children one or two years younger or older also be of interest. What about a study of sixth graders?

Let's say the specific circumstance that prompted this clinician's search for evidence was the arrival on his/her caseload of a girl who just arrived in the U.S., from China, and has limited English proficiency. Should the evidence that is sought be limited to English-language learners, or should children whose first language is English also be included?

Intervention and Comparison

The intervention of interest is specified as individual pull-out treatment, and the comparison group pull-out treatment. Here the decision has to be made whether that is specific enough, i.e., whether more information is needed about what actually transpires during those sessions, or perhaps whether it is too specific, and the real issue is whether the child should be treated using a pull-out model (whether individual or group) as opposed to a classroom-based or collaborative consultation approach.


Finally, the outcome of interest as written here is the child's ability to pronounce /s/. Another approach would be to look at more functional outcomes such as the extent to which the teacher felt that the child was better able to participate in classroom activities requiring speech. The two different types of outcomes would likely involve very different types of evidence (for more on this topic, see Robey R (2004). A five-phase model for clinical-outcome research. Journal of Communication Disorders, 37, 401-411).

How to make these choices? There are two factors that will influence the final question. The first is the theoretical model that informs the clinician's thinking about this disorder. It is the clinician's expertise that will dictate whether he or she thinks it is important to differentiate between left- and right-hemisphere strokes for a particular question, or to what extent the age of a patient should be considered a factor. This is one important reason that true EBP must involve the expertise and experience of the clinician.

Once these initial decisions are made, the second factor comes into play, and that is the availability of evidence. This serves as a reality check on the feasibility of garnering sufficient evidence to address the question as posed. Perhaps, the question was originally posed as elementary-school students with articulation disorders, but it turns out that there is a large body of evidence specific to Kindergartners and/or specific to English-language learners, so that the question can be more focused, and thus more directly relevant to the particular clinical circumstance that prompted it. More common, however, is a dearth of evidence, and the clinician is faced with using his or her own expertise to decide to what extent the criteria can be relaxed (e.g., going from evidence on "left-hemisphere stroke patients" to just stroke patients) while still maintaining relevance.

For more on framing clinical questions, see The University of Washington's web-based tutorial, Constructing Well-built Clinical Questions using PICO.

Next Step: Finding the Evidence

ASHA Corporate Partners