Division of Laboratory Sciences, National Center for Environmental Health, Centers for Disease Control and Prevention, Atlanta, Georgia, USA.
J Expo Sci Environ Epidemiol. 2010 Jan;20(1):29-37. doi: 10.1038/jes.2008.72. Epub 2008 Nov 12.
Biomonitoring involves the assessment of human or animal populations by measuring organic or biological compounds or their metabolites in the body fluids or tissues of individuals in those populations. Pooling samples before making analytical measurements can reduce the costs of biomonitoring by reducing the number of analyses. By proper choice of pooled-sample design, population means can be estimated without measuring individual samples. I present a statistical method for characterizing an entire population distribution of such compounds by exploiting the theoretic relationship between interindividual-sample variance and the variation between pooled samples. I use simulation experiments to determine an optimum pooled-sample design as a function of the number of subpopulations and the number of available samples. Using pooled samples to characterize populations is not only more cost-efficient, but also in some cases it can lead to more precise and less biased parameter estimation than that occurs with individual samples.
生物监测是通过测量人体或动物群体中个体的体液或组织中的有机或生物化合物或其代谢物来评估这些群体。在进行分析测量之前，对样本进行汇集可以通过减少分析的数量来降低生物监测的成本。通过适当选择汇集样本设计，可以在不测量个体样本的情况下估计群体平均值。我提出了一种通过利用个体样本间方差与汇集样本间变异之间的理论关系来描述此类化合物整个群体分布的统计方法。我使用模拟实验来确定作为亚群数量和可用样本数量函数的最佳汇集样本设计。使用汇集样本来描述群体不仅更具成本效益，而且在某些情况下，它可以导致比使用个体样本更精确和偏差更小的参数估计。
