Bailey K R
Stat Med. 1987 Apr-May;6(3):351-60. doi: 10.1002/sim.4780060327.
In determining the role inter-study variation should play in an overview analysis, it is important to consider three factors: which question one is trying to answer; the degree of similarity or dissimilarity of design, and the degree to which heterogeneity of outcomes can be explained. Three questions one might be interested in are: whether treatment can be effective in some circumstances; whether treatment is effective on average, and whether treatment was effective on average in the trials at hand. Under the assumption of no qualitative interaction, the answers to these questions coincide. The O-E analysis most directly answers the third question. Other analyses are suggested when the first question is of interest, using the aspirin post-MI studies as an example.
在确定研究间变异在概述分析中应发挥的作用时，考虑三个因素很重要：一是试图回答的问题；二是设计的相似或不同程度；三是结果异质性能够得到解释的程度。可能感兴趣的三个问题是：治疗在某些情况下是否有效；治疗平均而言是否有效；以及治疗在手头的试验中平均而言是否有效。在不存在定性交互作用的假设下，这些问题的答案是一致的。观察值与期望值（O-E）分析最直接地回答了第三个问题。当对第一个问题感兴趣时，以心肌梗死后阿司匹林的研究为例，建议采用其他分析方法。
