如何正确处理单因素多水平设计定量资料——怎样在药物应用与监测研究中正确运用统计学（二）

现代科研中，许多人希望同时考察多个实验组所代表的总体均值之间的差异是否有统计学意义，科研工作者们处理这类资料最常用的统计分析方法就是t检验和单因素多水平设计定量资料方差分析，而这些应用中相当一部分是误用或滥用。其实，正确处理单因素多水平设计定量资料，关键就在于能够正确地辨析出实验设计类型和定量资料是否满足参数检验前提条件。本文以实例为基础，讨论如何正确处理单因素多水平设计定量资料。     

均匀试验设计在复方碳酸钙泡腾颗粒剂处方筛选中的应用

以钙离子浓度为指标 ,通过单因素考察和均匀设计法筛选出泡腾剂最佳处方组成 ,结果表明 ,均匀设计是进行多因素多水平条件优化的理想实验设计方法。     

用SAS软件实现具有一个重复测量的单因素设计定量资料的统计分析

重复测量设计是药学科研临床试验中经常用到的一种试验设计方法,在临床药物疗效和安全性评价研究方面具有十分广泛的适用性。本文主要从具有一个重复测量单因素设计的概念,统计分析方法的合理选用,如何使用SAS程序进行分析,以及如何解释结果几个方面作一概述。     

均匀设计及其在药学科研设计中的应用

本文阐明如何使用均匀设计来解决药学领域中的多因素多水平的试验设计问题。均匀设计的最大优点是：试验的次数少，每个因素每个水平只做一次试验，且试验次数与水平数相等。而正交设计的试验次数则是水平数平方的整数掊。如果我们将均匀设计和多元统计相配合作综合分析，更利于分析各因素对试验结果的影响，定量地预测优化条件。所以均匀设计是值得在药学科研领域中推广应用的一种好的设计方法。     

如何正确运用单因素方差分析——药物研究中的统计学（二）

本文以案例为基础,介绍单因素方差分析基本概念、软件操作与结果解读,讨论如何正确应用单因素方差分析处理平行设计的随机对照试验的定量资料。     

直肠癌205例预后影响因素分析

目的 探讨直肠癌患者预后的影响因素.方法 选择205例直肠癌患者进行单因素及多因素分析,以了解直肠癌的临床及病理因素与预后的关系.结果 单因素分析显示,肿瘤组织学类型、肿瘤浸润深度、淋巴结转移、远处转移、手术方式及术前CEA和CA199水平与预后相关.多因素分析显示,肿瘤的浸润深度、淋巴结转移、远处转移、手术方式及术前CEA水平是影响直肠癌预后的独立因素.结论 病理分期是决定直肠癌患者预后的最重要因素,而手术方式和术前CEA水平可以作为判断直肠癌患者预后的重要参考因素.     

射血分数保留的心力衰竭患者血尿酸水平与心房颤动的关系

目的探讨射血分数保留的心力衰竭患者血尿酸水平与心房颤动发生的关系。方法连续入选心力衰竭患者150例,通过单因素方差分析各组间血尿酸浓度的差异,多因素Logistic回归分析房颤和血尿酸等相关因素的关系。结果三组患者在年龄、性别、空腹血糖、总胆固醇、三酰甘油、高密度脂蛋白、低密度脂蛋白、尿素氮、血肌酐等方面无显著差异。多因素Logistic回归分析显示,尿酸和左心房大小与房颤有显著相关性。结论射血分数保留的心力衰竭患者的血尿酸水平与心房颤动的发生相关,血尿酸越高越容易发生持续性房颤。     

药材玉竹RAPD反应体系的优化

目的寻找玉竹的最佳RAPD反应体系，保证实验结果的稳定可靠性。方法采用单因素多水平的梯度实验和正交设计实验相结合，对两种不同实验方法所得结果进行分析。结果确定了适合于玉竹的最佳RAPD反应体系。结论优化的RAPD反应体系是保证实验结果稳定可靠的重要因素之一。     

直肠癌伴同时性肝转移危险因素分析

目的探讨直肠癌同时性肝转移患者与临床病理因素间关系。方法收集安徽医科大学第一附属医院2006年1月1日～2009年12月31日直肠癌患者359例。回顾性分析直肠癌肝转移患者与临床病理因素关系。选择性别、年龄、血清CEA水平、血清CA19-9水平、病程、大体类型、组织类型、肠壁侵袭深度、病理淋巴结转移和肿瘤大小等10个可能影响肝转移的因素,单因素用χ2检验及t检验,多因素用Logistic回归(SPSS17.0统计软件)进行分析。结果单因素分析显示,年龄(P=0.037)、血清CEA水平(P=0.000)、肠壁侵袭深度(P=0.022)、病理淋巴结转移(P=0.000)与直肠癌同时性肝转移有关。多因素分析显示仅血清CEA水平及病理淋巴结转移与肝转移相关,无论是单因素、还是多因素回归分析,性别、血清CA19-9水平、大体类型、组织类型、肿瘤大小均与肝转移无关。结论血清CEA水平及病理淋巴结转移是直肠癌伴同时性肝转移的2个显著性危险因素。性别、血清CA19-9水平、病程、大体类型、组织类型及肿瘤大小与肝转移无关。     

维持性血液透析患者继发性甲状旁腺功能亢进相关因素分析

目的:探讨影响维持性血液透析患者继发性SHPT的相关因素。方法:收集接受维持性血液透析治疗的患者98例,对其临床资料进行调查记录,利用单因素分析和多因素Logtic回归分析对于患者继发性SHPT有关的因素进行分析。结果:98例患者中有继发性SHPT47例,单因素与分析发现该病受年龄、血透时间、血红蛋白水平、血钙水平、血磷水平、肌酐水平、C反应蛋白、钙磷乘积等影响较大;多因素Logtic回归分析证实透析时间、血磷、C反应蛋白及肌酐为该病的独立危险因素。结论:维持性血液透析患者继发性SHPT受患者的年龄、血透时间及多种生化指标影响,其中透析时间、血磷、C反应蛋白及肌酐可以作为该病的独立危险因素。     

多种统计学优化方法在盐酸川芎嗪缓释制剂处方设计中的比较研究

处方优化方法的合理选用对高效、经济地设计药物剂型发挥着很大的作用。目前在药剂学研究中普遍应用的统计学实验设计方法有正交设计、均匀设计、析因设计、单纯形设计及中央多点等距设计(CCD)等。本文以盐酸川芎嗪(TMPH)缓释片剂为模型药物,通过其优化设计过程,对多种优化方法的设计要求和预测能力进行了横向对比,期望能有助于处方优化方法的合理选择。     

Ⅱ类骨质量下短种植体表面设计对骨界面应力分布的影响

目的：探讨骨质量和种植体表面设计对短种植体-骨界面应力分布的影响。方法：利用3D-Doctor软件和ABAQUS软件建立Ⅴ形螺旋设计、反支撑形螺旋设计、支撑形螺旋设计和鳍式非螺旋设计的短种植体植于Ⅰ类骨质量的上颌骨后区的三维有限元法模型4个。对所有模型进行垂直和侧向加载，分析比较周围骨组织的应力分布和Von-Mises应力峰值。结果：垂直向和侧向加载时，反支撑形和支撑形螺纹设计与鳍式非螺旋设计都表现出较好的应力分布特征。结论：支撑形螺旋设计、反支撑形螺旋设计和鳍式非螺旋设计均适用于短种植体，应避免应用Ⅴ形螺旋表面设计。     

临床试验Williams设计中样本含量的估算方法

在临床试验中,2组交叉设计应用已相当广泛,其样本含量估算方法也被研究者所熟悉。多组交叉试验由Williams首先提出,因此被称为Wil-liams设计。本文介绍基于Williams设计的样本含量估算方法,并提供示例分析,供研究者参考使用。     

Design of population pharmacokinetic experiments using prior information

A population pharmacokinetic study design is a group of elementary designs each composed of a set of sampling times to be performed in a number of subjects in the design. Design factors such as the number of elementary designs, proportion of subjects in each elementary design, number of samples per subject, and sampling times in the subjects need to be carefully balanced in the design of a study. An optimally designed population pharmacokinetic study will give the best combination of these design factors and involves application of statistical experimental design principles to non-linear population pharmacokinetic models. Information from previous experiments, the literature, and experience in the form of model and parameter estimates are used to design a future study by optimization of a design criterion within some constraints. This work provides a brief background of approaches to the designs of a population pharmacokinetic experiment and a review of optimal design methodologies that have been developed for designing population pharmacokinetic experiments. Computer application programs and software that have been developed together with options available in them are also reviewed.     

Design of small molecule libraries for NMR screening and other applications in drug discovery

There are conceptual differences between high-throughput screening (HTS) and fragment-based screening by NMR. The number of compounds in libraries for NMR screening may be significantly smaller than those used for HTS. Because one relies on a small library its design is significantly important and is the object of this article. A short introduction on fragment-based NMR screening approaches will be provided. Although there are currently very few reports describing the design of libraries of small molecules for NMR screening, aspects of the question of how to compile diverse collections of small molecular fragments useful for drug design were previously addressed for the purposes of combinatorial library design and de novo drug design. As these disciplines are highly interrelated and are applied in an interconnected manner with NMR screening within the drug discovery process, a review of combinatorial library design and especially the building block or fragment selection strategies applied for combinatorial library design and de novo design is well suited to reveal fundamental strategies and potential techniques for the design of NMR screening libraries. This section will be rounded off by a report on hands-on-experience with the design of the Novartis second-site NMR screening library and practical considerations for the design of compound mixtures. Rather than providing an exact protocol general guidelines will be indicated.     

大数据和人工智能技术用于计算机辅助药物设计的研究进展

大数据和人工智能（AI）技术不仅可以对海量的生物医学数据进行准确和综合地分析,而且可以帮助构建药物设计领域的预测模型。随着算法和统计方法的发展,大数据和AI技术已应用于计算机辅助药物设计（CADD）。CADD可被用于有效克服药物设计领域的困难,从而高效地设计和开发新药。介绍了药物设计和发现过程中数据预处理和建模步骤、药物设计和发现过程中基于AI的建模方法、大数据和AI技术在CADD中的最新应用,以期为我国药物设计和开发提供参考。     

微球给药系统的工艺优化研究

目的 :阐述微球制备工艺中的优化技术。方法 :检索近年来国内外文献并进行整理和归纳。结果 :单因素设计、正交设计、均匀设计、因子设计是目前微球工艺中常用的优化方法。结论 :通过上述方法优化微球的制备工艺 ,可在一定条件下达到经济、高效的目的 ,更理想的优化技术尚有待研究     

均匀设计方法及其应用

利用统计学方法处理科学试验数据是当今在科学研究中重要的一个环节 ,但应当注意到在试验过程中试验设计是否合理直接影响到试验的结果和统计分析。另一方面 ,科学的试验设计可以大量的节省试验次数 ,达到试验的最佳效果。通过对均匀设计与正交设计的比较 ,说明了均匀设计可用较少的试验次数达到较好的试验效果 ,同时指出了其在应用中应注意的问题。     

A Bayesian Approach to the ICH Q8 Definition of Design Space

Manufacturers of pharmaceuticals and biopharmaceuticals are facing increased regulatory pressure to understand how their manufacturing processes work and to be able to quantify the reliability and robustness of their manufacturing processes. In particular, the ICH Q8 guidance has introduced the concept of design space. The ICH Q8 defines design space as “the multidimensional combination and interaction of input variables (e.g., material attributes) and process parameters that have been demonstrated to provide assurance of quality.” However, relatively little has been put forth to date on how to construct a design space from data composed of such variables. This study presents a Bayesian approach to design space based upon a type of credible region first appearing in Peterson's work.This study considers the issues of constructing a Bayesian design space, design space reliability, the inclusion of process noise variables, and utilization of prior information, as well as an outline for organizing information about a design space so that manufacturing engineers can make informed changes as may be needed within the design space.     

Design of population pharmacokinetic experiments using prior information

A population pharmacokinetic study design is a group of elementary designs each composed of a set of sampling times to be performed in a number of subjects in the design. Design factors such as the number of elementary designs, proportion of subjects in each elementary design, number of samples per subject, and sampling times in the subjects need to be carefully balanced in the design of a study. An optimally designed population pharmacokinetic study will give the best combination of these design factors and involves application of statistical experimental design principles to non-linear population pharmacokinetic models. Information from previous experiments, the literature, and experience in the form of model and parameter estimates are used to design a future study by optimization of a design criterion within some constraints. This work provides a brief background of approaches to the designs of a population pharmacokinetic experiment and a review of optimal design methodologies that have been developed for designing population pharmacokinetic experiments. Computer application programs and software that have been developed together with options available in them are also reviewed.