Festing Michael F W
Medical Research Council Toxicology Unit, Leicester, United Kingdom
Toxicol Pathol. 2014 Dec;42(8):1238-49. doi: 10.1177/0192623313517771. Epub 2014 Jan 31.
The results of repeat-dose toxicity tests are usually presented as tables of means and standard deviations (SDs), with an indication of statistical significance for each biomarker. Interpretation is based mainly on the pattern of statistical significance rather than the magnitude of any response. Multiple statistical testing of many biomarkers leads to false-positive results and, with the exception of growth data, few graphical methods for showing the results are available. By converting means and SDs to standardized effect sizes, a range of graphical techniques including dot plots, line plots, box plots, and quantile-quantile plots become available to show the patterns of response. A bootstrap statistical test involving all biomarkers is proposed to compare the magnitudes of the response between treated groups. These methods are proposed as an extension rather than an alternative to current statistical analyses. They can be applied to published work retrospectively, as all that is required is tables of means and SDs. The methods are illustrated using published articles, where the results range from strong positive to completely negative responses to the test substances.
重复剂量毒性试验的结果通常以均值和标准差（SDs）表格形式呈现，并对每个生物标志物给出统计学显著性的说明。解释主要基于统计学显著性模式，而非任何反应的幅度。对众多生物标志物进行多次统计检验会导致假阳性结果，并且除生长数据外，几乎没有用于展示结果的图形方法。通过将均值和标准差转换为标准化效应量，一系列图形技术（包括点图、线图、箱线图和分位数 - 分位数图）可用于展示反应模式。提出了一种涉及所有生物标志物的自助统计检验，以比较处理组之间反应的幅度。这些方法被提议作为当前统计分析的扩展而非替代方法。它们可以追溯应用于已发表的研究，因为所需的全部内容只是均值和标准差表格。使用已发表的文章对这些方法进行了说明，其中对受试物质的结果范围从强阳性到完全阴性反应。
