Clinical trials can be designed to determine the efficacy and safety of a specific intervention for fatigue. As noted, this may be drug, such a psychostimulant or a corticosteroid, or any of a variety of non-pharmacologic approaches.

Selecting a Design

Study design may be viewed as a hierarchy of progressive controls, the purpose of which is to reduce the risk of systematic bias that could lead to erroneous results. The likelihood that the experimental findings are valid increases as greater experimental control is exercised.

Quasi-experimental studies. The least control, and hence the greatest doubt about validity of the findings, occurs when an intervention for fatigue is administered openly to a series of patients without comparison data. This approach is really a type of survey that can provide information about safety, and some information about efficacy, but must be interpreted cautiously. There are potentially powerful effects that could influence response and preclude attribution of the outcomes to the study intervention. These include the placebo effect and a tendency for subjects entering a clinical trial to "regress to the mean." In surveys of fatigue, the latter effect presumably occurs because many subjects agree to participate in a trial when the problem is relatively severe and the desire for help is greatest. From this condition, there may be a tendency to improve simply as a matter of the natural ebb and flow of the symptom. This naturally occurring improvement could be misinterpreted as an effective therapeutic response to an otherwise useless intervention.