Statistical Methods In Epidemiology--general

该项目为流行病学开发新的统计方法，具有广泛的应用，并为正在进行的流行病学项目，特别是与生殖研究有关的项目开发所需的方法。 今年的工作涉及几个项目。(1)一个项目涉及在评估围产期结果的风险因素时对出生时胎龄进行统计调整的常见做法。 由于出生时的胎龄可能由其他因素引起，这些因素也会影响结果，因此它是一个“碰撞器”，这种调整可能会在估计中产生偏差。 我们进行的计算表明，在合理的假设下，这种偏差可能是实质性的，因此生殖流行病学家应该避免这种调整模型的病因推断。(2)另一个项目涉及基于人类样本的昂贵生物标志物的汇总评估。早期的工作表明，在病例对照设置中，可以将病例组和对照组的样本汇集在一起，并进行基于集合的分析。 通过略微修改的逻辑模型，该分析可以估计个体水平的风险参数，并且与基于个体测定的分析相比几乎没有损失功效。 这意味着，如果暴露是基于使用人体样本的昂贵测定，则可以通过在测定前合并样本来显著提高效率。在最近的工作中，我们已经将这些方法扩展到适用于精细匹配的病例对照设计和妊娠时间数据。 在后一种设计中，人们将样本集中在由受孕时间定义的地层中。 同样，与单个水平测定相比，功率几乎完全不受影响，并且成本大大降低。一般来说，共享的另一个巨大好处是保守地使用不可替代的人类标本。例如，将样本合并为3个样本组可将每个样本所需的样本量减少2/3。我们现在正在开发时间-事件设置中的“前瞻性汇总”方法。 这种方法应能对环境暴露对生存和其他故障时间结果的影响进行有力和廉价的评估。 (3)第三个项目开发了改进的方法，用于评估基于生物标志物的纵向测量与假阳性和假阴性预测相关的未来事件预测的准确性。 一份基于精细匹配病例对照研究中合并暴露评估的设计和分析方法的论文正在出版，一份提出并评估在环境因素对生育力影响的妊娠时间研究中使用合并的论文正在进行第二次审查。 一篇基于在涉及生物标志物的纵向研究中评估预测的论文发表在《生物统计学》上。

This project develops new statistical methods for epidemiology with broad applications and also methods as needed for ongoing projects in epidemiology, particularly those related to reproductive studies. The work this year involved several projects. (1) One project concerned the common practice of statistically adjusting for gestational age at birth when assessing risk factors for perinatal outcomes. Since gestational age at birth can be caused by other factors that also influence the outcome, it is a "collider" and such adjustment can potentially produce bias in the estimation. We carried out calculations to show that under reasonable assumptions such bias can be substantial and thus reproductive epidemiologists should avoid such adjusted models for etiologic inference. (2) Another project concerns pooled assessment of expensive-to-assay biomarkers based on human samples. Earlier work had shown that in a case-control setting one can pool together specimens from sets of cases and sets of controls and carry out a set-based analysis. With a slightly modified logistic model that analysis can estimate the individual-level risk parameters and loses almost no power compared to analysis based on individual assays. This means that if an exposure is based on an expensive assay that uses human samples, one can markedly improve efficiency by pooling specimens prior to assay. In recent work, we have extended these methods to apply to a fine-matched case-control design and also to time-to-pregnancy data. With the latter design, one pools specimens within strata defined by the time to conception. Again the power suffers almost not at all, compared to individual level assays, and the costs are greatly reduced. Another great benefit to pooling in general is in its conservative use of irreplaceable human specimens. For example, pooling specimens in sets of 3 reduces the amount needed from each specimen by 2/3. We are now developing methods for "prospective pooling" in time-to-event settings. Such methods should allow powerful and inexpensive assessments of the effects of environmental exposures on survival and other failure-time outcomes. (3) A third project developed improved methods for assessing the accuracy of prediction of future events based on longitudinal measures of a biomarker in relation to false positive and false negative predictions. A paper based on methods for design and analysis of pooled exposure assessments in fine-matched case-control studies is in press, and one that proposes and evaluates the use of pooling in time-to-pregnancy studies of fertility effects of environmental factors is undergoing its second review. A paper based on assessing prediction in a longitudinal study involving a biomarker was published in Biometrics.