After the outcome question is refined and the pertinent literature is reviewed, but before the data collection process is designed, it is important to identify the rationale for conducting the study. The clinician should be clear about why the study should be conducted and how the results of the investigation will be used in practice. Outcome studies, even when limited to one clinician’s patient care practice, require an investment of time and energy. This chapter will describe the reasons for identifying a rationale and how to integrate it with the outcome question.
THE RATIONALE: AN OPERATIONAL DEFINITION
The rationale explains the purpose for conducting the study. The rationale includes a hypothesis about the relationship among the selected inputs, processes, and outcomes and how the potential results will be applied in practice. (See Fig. 8-1 for an example of a rationale.) Regardless of whether the rationale is based on theory or practical factors, it should provide an explanation for the choice of characteristics and interventions to study as well as describe how the study results will improve service delivery.
Example of an outcome question, a rationale, and the predicted use of data.
Why a Rationale Is Needed
Outcome studies should not be started unless the clinician can answer the question, “How will the results be used?” The rationale puts the study in perspective and guides the clinician in the use of the results.
Do not start a study without knowing how an answer will influence clinical practice.
There are three possible results for most comparative outcome studies. The results may show that
“Variable A” is better than “Variable B”
“Variable B” is better than “Variable A”
“Variable A” is equivalent to “Variable B”
In descriptive studies, in which the clinician identifies the incidence of a particular variable or process, there are four possible answers. The data may show
Strong evidence for the variable or process
Strong evidence against the variable or process
The variable or process appears equivalent to other variables or processes
No support for the variable or process
In a well-considered rationale, there is a plan of action for each of the possible outcomes. Thus, in a comparison study, if variable A appears better than variable B, the clinician might plan to use A more often. If B appears better than A, then the opposite might occur. If there is no difference between A and B, the clinician might choose randomly ...