Group Testing in the Presence of Error with Application to HIV/AIDS

描述(由申请人提供)：汇集样本以提高筛查个人罕见疾病的效率由来已久，可以追溯到20世纪40年代在新兵中筛查梅毒。多年来，标本池或群体测试已被应用于昆虫学、遗传学、国际贸易、制药和血库行业、计算机科学、分析化学和许多其他领域。在传染病方面，群体检测通常用于(I)病例识别，即检测所有患有相关疾病的个人，以及(Ii)流行率估计，即估计个人在人群中患有特定疾病的比例。作为前者的一个例子，目前美国的血库和北卡罗来纳州的公共艾滋病毒检测系统使用样本池来检测最近感染艾滋病毒的个人。集体检测可能比单独检测更有效、准确和精确；然而，由于大量和多样化的集体检测文献，许多支持使用集体检测的结果在传染病环境中没有得到承认。我们建议研究群体测试的统计学方面，并将其应用于艾滋病毒/艾滋病和其他传染病的病例识别和流行率估计。我们的具体目标是推导和比较多阶段和基于阵列的分组测试算法在存在测试误差的情况下的识别和估计问题的操作特性。测试误差模型将考虑恒定和池大小相关的灵敏度和特异度。我们假设阵列测试算法在效率和精度方面优于传统的多级测试算法。虽然该方法是以检测急性艾滋病毒为基础的，但我们的结果将直接适用于其他传染病。

DESCRIPTION (provided by applicant): Pooling of specimens to increase efficiency of screening individuals for rare diseases has a long history, dating back to screening of syphilis in military inductees in the 1940s. Over the years, specimen pooling or group testing has been applied in entomology, genetics, international trade, the pharmaceutical and blood bank industries, computer science, analytical chemistry, and many other areas. In the context of infectious diseases, group testing is typically used for (i) case identification, i.e., detecting all individuals having the disease of interest, and (ii) prevalence estimation, i.e., estimating the proportion of individuals in the population having a particular disease. As an example of the former, currently the nation's blood banks and the state of North Carolina's public HIV testing system employ specimen pooling to detect individuals recently infected with HIV. Group testing can be more efficient, accurate, and precise than individual testing; yet, many of the results supporting the use of group testing have gone unrecognized in the infectious disease setting owing to the vast and diverse group testing literature. We propose to research statistical aspects of group testing with application to case identification and prevalence estimation in HIV/AIDS and other infectious diseases. Our specific aims are to derive and compare the operating characteristics of multistage and array based group testing algorithms for both the identification and estimation problems in the presence of test error. Models for test error will consider constant and pool size dependent sensitivity and specificity. We hypothesize that array testing algorithms are superior to conventional multistage testing algorithms in terms of efficiency and accuracy. While the methodology is motivated by detection of acute HIV, our results will be directly applicable to other infectious diseases.