Cross-over method

INCORRECT

The correct answer is (b).

The cross-over is associated with additional risk of bias. There is no information about the duration or nature of drug effects after dosing is stopped. If positive effects were prolonged, or patients tended to become more fatigued, or more depressed, after treatment with active drug was stopped, carryover effects could preclude interpretation of the second study phase. Period effects (e.g. outcomes observed during the first phase of the study may be relatively enhanced by patient expectation) and time-dependent effects (e.g. after several weeks, regression to the mean occurs and fatigue naturally lessens in some patients) could also compromise the interpretation of the data from the second phase of a two phase cross-over trial. More sophisticated, multiple phase cross-over studies could be designed to address these concerns, but these do not eliminate the concern about carryover effect and add too much to respondent burden. The simplest approach, therefore, is a parallel group design.