Sugita M, Yamaguchi N, Izuno T, Kanamori M, Kasuga H
Department of Environmental and Occupational Health, Toho University School of Medicine, Tokyo, Japan.
Tokai J Exp Clin Med. 1994 Sep;19(1-2):29-37.
A summarized odds ratio, calculated from odds ratios of published studies in a meta-analysis, may be overestimated because of publication bias. A method has been developed estimating indirectly the summarized odds ratio of all studies in a given research area, including not only those published but also unpublished. In the present study, a publication probability according to odds ratio value was obtained from the probability density function of all the studies, and a histogram of those published. A publication probability, according to odds ratio value, enables us to infer the quantitative relationship between publication probability and odds ratio. A notable nonpublication of studies whose odds ratios were close to unity was shown from examples of studies on the relationship between passive smoking and lung cancer, whereas nonpublication of studies whose summarized odds ratio was located far from unity was not detected from examples of studies on the relationship between cryptorchidism and testicular cancer. In both study areas, however, the differences between the summarized odds ratios, either with or without the hypothetical unpublished studies, were not large. The small difference, however, should not be ignored, when the study area is recognized as a social problem.
在荟萃分析中，根据已发表研究的比值比计算得出的汇总比值比可能会因发表偏倚而被高估。已经开发出一种方法来间接估计给定研究领域中所有研究的汇总比值比，这些研究不仅包括已发表的，还包括未发表的。在本研究中，根据所有研究的概率密度函数以及已发表研究的直方图，获得了根据比值比值得出的发表概率。根据比值比值得出的发表概率使我们能够推断发表概率与比值比之间的定量关系。从被动吸烟与肺癌关系的研究实例中可以看出，比值比接近1的研究存在明显的未发表情况，而从隐睾症与睾丸癌关系的研究实例中未发现汇总比值比远离1的研究存在未发表情况。然而，在这两个研究领域中，无论是否包含假设的未发表研究，汇总比值比之间的差异都不大。然而，当该研究领域被视为一个社会问题时，这种小差异不应被忽视。
