Einarson T R, McGhan W F, Bootman J L, Sabers D L
Am J Hosp Pharm. 1985 Sep;42(9):1957-64.
Meta-analysis, a quantitative method of combining the results of independent research studies, is described as a method for reviewing research literature. Four steps are taken to summarize the research in an area. First, a thorough literature review is conducted to identify a group of research studies with the relevant treatment variable. Second, an effect size is calculated for each study. Third, an overall (composite) effect size is determined by a weighted combination of the obtained effect sizes. Fourth, a fail-safe N (the number of unpublished studies with opposing conclusions needed to negate the published literature) is calculated to assess the certainty of the overall effect size. Meta-analysis was applied to 33 studies of chymopapain in an illustrative example. The analysis produced a large effect size of 0.8082 and a fail-safe N of 214, indicating strong support for the effectiveness of the treatment with chymopapain. Meta-analysis can be a useful tool if it is used properly. It is particularly useful as an adjunct to other methods of review that are used in pharmacy practice.
元分析是一种整合独立研究结果的定量方法，被描述为一种回顾研究文献的方法。总结某一领域的研究需采取四个步骤。首先，进行全面的文献综述以确定一组具有相关治疗变量的研究。其次，为每项研究计算效应量。第三，通过对所得效应量进行加权组合来确定总体（综合）效应量。第四，计算失效安全数N（即需要多少未发表的、得出相反结论的研究才能否定已发表的文献）以评估总体效应量的确定性。在一个示例中，对33项关于木瓜凝乳蛋白酶的研究应用了元分析。分析得出的效应量为0.8082，失效安全数N为214，表明对木瓜凝乳蛋白酶治疗效果有有力支持。如果使用得当，元分析可以成为一个有用的工具。它作为药学实践中使用的其他综述方法的辅助手段尤其有用。
