Lewis J A
ICI Pharmaceuticals Division, Macclesfield, Cheshire, UK.
Neuroepidemiology. 1987;6(4):198-208. doi: 10.1159/000110120.
This paper evaluates the relative merits of the crossover design and the parallel group design for the evaluation of migraine therapy, focussing on prophylaxis. Three issues are highlighted which lead to a preference for the parallel group trial: changes in the treatment effect with time, the possibility of carry-over effects, and the handling of withdrawals and protocol deviations. These have to be weighed against the greater power of the crossover design. It is shown that the difference in power is relatively small when baseline or run-in assessments are available. It is therefore concluded that the parallel group trial is the design of choice for prophylaxis, and that it has a place in the evaluation of acute therapy.
本文评估了交叉设计和平行组设计在偏头痛治疗评估（重点是预防性治疗）中的相对优点。突出了三个导致更倾向于平行组试验的问题：治疗效果随时间的变化、残留效应的可能性以及对退出和方案偏离的处理。这些必须与交叉设计更大的效能相权衡。结果表明，当有基线或导入期评估时，效能差异相对较小。因此得出结论，平行组试验是预防性治疗的首选设计，并且在急性治疗评估中也有一席之地。
