SEQUENTIAL ANALYSIS OF HEMATOLOGIC MEASUREMENTS

这项拟议的研究将开发序列数据的统计方法 血液学检测结果分析为患者提供特异性参考 有助于检测和诊断贫血的价值。 血液学系列检查结果的自动存储和分析 现在用现在的实验室仪器在技术上是可行的 用于数据存储和处理的设备。在目前的实践中，医生们 在头脑中将实验室结果与之前的值进行比较，并使用他们的 临床判断，以确定任何变化的意义。这个 拟议的研究旨在通过开发 按顺序分析测试结果并识别任何偏离的方法 从一个病人的过去的价值观。虽然以人口为基础 参考范围现在用于评估单个测试值，使用 从先前对单个患者的研究中得出的参考值 将提供一种更敏感的检测异常的手段，特别是 在患有慢性病的患者中。以证明其临床应用价值 这种方法，本项目将重点研究检测和评估 血液透析患者血红蛋白检测中的贫血 浓度、红细胞压积、红细胞计数、平均红细胞体积和红色 细胞体积分布。特别是我们的初步经验， 表明对红细胞体积分布的顺序分析将会 提供早期和灵敏的小细胞或大细胞疾病的指征 红细胞平均体积改变前的红细胞生成或 血红蛋白浓度。 得出特定患者的统计方法的发展 此组的参考值将使自动检查成为可能 快速、可靠的血液透析装置的实验室数据 血液学检测显示患者的身份 与过去的价值观有很大的不同。提供有效的统计数据 从序列中导出特定于患者的参考范围的方法 血液学测量，这个研究项目将有三个具体的 目标：(1)开发和验证序列数据的统计方法 分析个体血液学测量以提供患者特异性 血红蛋白浓度、红细胞压积、红细胞参考值 计数和平均红细胞体积；(2)建立和验证统计学 血液学分布的序贯分析方法 提供针对患者的红细胞体积比较的测量方法 分布；(3)确定临床应用的可行性和实用性 血液学测量的序贯分析用于评估 一组慢性肾衰竭患者，他们接受了 血液透析。尽管在一系列有限的实验室测量中 在拟议的研究中，特定的患者组已被选中进行研究 对于项目而言，将得出的方法将具有普遍适用性。 为实验室自动化评审提供统计基础 使用特定于患者的参考值的数据应有助于评估 通过使医生能够及早、敏感和 可靠地识别与每一个的过去值相比的重大变化 有耐心的。

The proposed research will develop statistical methods for the sequential analysis of hematologic measurements to provide patient-specific reference values that will facilitate the detection and diagnosis of anemia. Automated storage and analysis of the results of serial hematologic studies are now technically feasible with present-day laboratory instruments and devices for data storage and processing. In current practice, physicians mentally compare a laboratory result with previous values and use their clinical judgement to determine the significance of any change. The proposed research is designed to facilitate this process by developing methods to sequentially analyze test results and identify any departure from the past values of an individual patient. While population-based reference ranges are now used to evaluate individual test values, the use of reference values derived from previous studies of an individual patient would provide a more sensitive means of detecting abnormalities, especially in patients with chronic disorders. To demonstrate the clinical utility of this approach, this project will focus on the detection and evaluation of anemia in hemodialysis patients by examination of the hemoglobin concentration, hematocrit, red cell count, mean corpuscular volume and red cell volume distribution. In particular, our preliminary experience suggests that sequential analysis of the red cell volume distribution will provide an early and sensitive indication of micro- or macrocytic erythropoiesis that precedes alterations in the mean corpuscular volume or hemoglobin concentration. The development of statistical methods for deriving patient-specific reference values for this group would make possible automated examination of laboratory data for the hemodialysis unit with rapid and reliable identification of patients whose hematologic measurements have significantly changed from past values. To provide valid statistical methods for deriving patient-specific reference ranges from serial hematologic measurements, this research project will have three specific aims: (1) develop and validate statistical methods for the sequential analysis of individual hematologic measurements to provide patient-specific reference values for the hemoglobin concentration, hematocrit, red cell count, and mean corpuscular volume; (2) develop and validate statistical methods for the sequential analysis of distributions of hematologic measurements to provide patient-specific comparisons for red cell volume distributions; (3) determine the clinical feasibility and utility of using sequential analyses of hematologic measurements for the evaluation of a group of patients with chronic renal failure who are treated with hemodialysis. Although a limited series of laboratory measurements in a specific group of patients has been selected for study in the proposed project, the methods to be derived would be of general applicability. Providing the statistical foundation for the automated review of laboratory data using patient-specific reference values should facilitate evaluation of test results by physicians by making possible early, sensitive and reliable identification of significant changes from the past values of each patient.