两生存率比较的样本含量估计与检验效能估算

在审阅临床医学科研设计或专业论文工作中,还是常会遇见有关“样本含量估计”处理不当、含糊不清、甚至错误等问题。还是有部分研究者在临床医学科研设计或专业论文中,将“样本含量估计”问题笼统叙述、一带而过,似乎就可以应付临床医学科研设计或专业论文的生物统计学审稿工作者了。     

撰写医学期刊稿件的统计学参考

医学期刊上刊载的论文多数包含统计学分析。统计学方法的正确运用对医学科研论文的质量是至关重要的。然而,目前发表的许多论文在统计学方面存在着不同程度的缺陷。本文作者尝试为医学期刊稿件的作者们提供一个统计学方面的参考,希望能对提高医学期刊稿件的统计学水平有所裨益。 　　1　材料与方法部分 　　1.1　统计设计　 　　1.1.1　一般原则　投往医学期刊的论文多数是研究工作的总结、概括和升华。统计设计是提高论文质量的必要前提。只有缜密的统计设计才能引导出有说服力的研究结论。可以说,如果一篇稿件的统计设计不合理或存在明显缺陷,不必去阅读论文的其余部分,编辑们就有足够的理由不予采用。在材料与方法部分,作者应提供有关统计设计的所有必要信息。这里“必要”的标准是应使读者能够根据这些信息,按照作者所报告的作法（需保密的技术和数据除外）重复该项研究。这些信息应包括：统计研究设计的名称和主要做法;观察单位的类型、采用或排除的条件;样本来源与获取（如抽样）方法;观察单位分组方法;样本大小与为什么采用这样数量的样本;观察指标的类型与获取数据（观测）的方法;不同类型（调查或临床实验等）的研究,还需要一定的附加信息。     

关于医学统计学的解析

1　调查与实验研究设计科研设计的质量直接影响着实验结果的准确性、可靠性、严密性和代表性 ,是实验数据处理的前提 ,决定着科学研究的成败。一个完整的科研设计包括专业设计和统计设计。专业设计是指研究者对专业知识的把握能力 ,直接影响着实验的深度和水平 ;统计设计是指研究者对统计知识的正确应用 ,直接影响着科学实验的质量 ,两者互相结合 ,缺一不可。完整的研究设计主要包括 :立题、制订研究计划、明确设计 3要素 (研究对象、处理因素、实验效应 ) ,它们是保证实验成功的条件 (设对照、控制误差、确定样本及抽样方法、确定研究方式…     

医学科研设计时确定样本含量的方法（4）

医学科研设计时确定样本含量的方法（４）柯惠新，汤旦林（医学统计室）（上接１９９４；８（１）：４８）５估计优势比（ｏｄｄｓｒａｔｉｏ，ＯＲ）在进行病例对照研究时，常常需要计算优势比（也叫做比数比）。假定病例组和对照组与是否暴露于某危险因素的列联表如下：...     

样本量估计与检验效能分析（四）

样本含量估计是任何一项实验研究或调查研究都不可回避的问题。根据已知的基本情况和统计学知识,估计出合适的样本含量在科研实践中是十分有意义的。本文介绍了拟作成组设计定量资料与定性资料差异性检验时的样本含量估计方法。具体地说,就是仅针对成组设计且为差异性检验的情形,介绍了效应指标为定量资料和定性资料时估计样本含量的计算公式,以及基于公式和借用SAS软件中POWER过程分别实现样本含量估计的方法,并通过实例进行了讲解,对科研工作者在实验设计时正确落实重复原则具有很好的指导作用。     

肿瘤新药Ⅱ期临床试验常用二阶段设计的样本含量估计及其SAS实现

目的介绍肿瘤新药Ⅱ期临床试验常用Gehan 二阶段设计及Simon 二阶段设计的设计原理和样本含量估计问题,提供 SAS宏程序快速实现其样本含量估计。方法结合二项分布的确切概率原理说明单阶段设计、Gehan 二阶段设计及两种Simon二阶段设计样本含量计算过程。利用SAS宏程序,通过设置不同参数比较其优缺点。结果Simon极小极大设计与单阶段设计相比,不会增加最大样本含量;与Gehan 二阶段设计相比,保证了在实际有效率很低时早期终止试验。结论Simon 二阶段比单阶段设计和Gehan 二阶段设计更有优势,且极小极大设计在多数情况下比最优化设计受欢迎。     

药物临床试验设计的原则

药物研究是以人体为对象进行的研究，了解、评估药物的安全性和有效性，在对比的基础上评价药物的优劣，要求遵循药物临床试验规范（GCP）的原则，而药物临床试验设计是其核心。最近我们对一些杂志报道的药物临床试验报告166篇进行了分析研究，发现66篇出现统计误用，涉及试验设计方面就有74处，占126处错误的58．73％，主要是实验设计不完整，不严谨，甚至错误。由此可见此问题存在的普遍性。下面我们就有关药物临床试验设计的原则作一概述。药物临床试验设计包括专业设计和统计设计二部分。专业设计指从专业角度考虑实验的具体目标，选…     

样本量估计与检验效能分析（五）

样本含量与检验效能估计是研究设计的重要内容之一。本文介绍了拟作单组设计、配对设计和交叉设计定量与定性资料差异性检验时的样本含量和检验效能估计方法,具体地说,就是针对前述的3种设计且为差异性检验的情形,介绍了效应指标为定量和定性资料时估计样本含量和检验效能的计算公式,以及基于公式和借用SAS软件中的POWER过程分别实现样本含量和检验效能估计的方法,并通过实例进行了演示和讲解,对实际工作者在试验设计时正确落实重复原则具有很好的指导作用。     

一组计量资料的统计分析

很多研究都涉及数据,需要统计.需要统计就需要获得符合统计学要求的数据,就需要设计.设计取决于研究目的、研究条件,一定程度上也取决于对统计学的认识.熟悉统计学不仅有助于完成研究,而且可能从数据中发现或获取其他重要信息,从而改进原有研究,甚至引导新的研究、新的发现.     

医学统计工作的基本内容

按工作性质及其先后顺序，可将医学统计工作分为实验设计、收集资料、整理资料、分析资料。实验设计是开展某项医学研究工作的关键，包括医学专业设计和统计学设计，医学专业设计的内容包括研究对象纳入和排除标准、样本含量、获取样本的方法、分组原则、观察（检测）指标、统计方法等。收集资料的方法包括各种试验、检测或调查，要求资料完整、准确、及时、有足够数量、具有代表性和可比性等。整理资料包括原始资料的检查与核对、对资料进行分组与汇总等。分析资料即对资料进行统计学分析，包括进行统计描述和统计推断。     

单组设计一元定量资料的统计推断与实验设计（二）

In the former three issues of this periodicalf, we have introduced how to correctly express and describe univariate quantitative data of a singlegroup design as well as how to perform statistical inference. Still, there are some practical and crucial problems that need our attention： before the experimental data have been gained, how can we estimate a suitable sample size？ How much is the test power？ Thus, this article introduces the estimation of sample size and the test power in detail.     

单组设计一元定量资料的统计推断与实验设计(一)(英文)

In the former two issues of this periodical[1,2],we have introduced how to correctly express anddescribe univariate quantitative data of a single-groupdesign,including expressing the data by frequencydistribution tables, describing the data by somei mportant indexes,such as the mean,the discretescale,the maxi mum and the mini mum values, ect .However, we do not do researches for the sample,but rather for the population. We expect to getmore information about the population from thesample data.This is called...     

如何表达与描述定量资料（三）

在医学论文中,如何正确表达与描述定量资料,对于正确选择统计学中的假设检验方法至关重要。当试验涉及两个、三个或多个因素,且这些因素之间依据专业知识和重复试验次数多少认为可以考察其各种关系时,该定量资料就属于可以考察因素之间交互作用的定量资料。根据试验设计类型和专业知识,运用恰当的方法表达和描述可以考察因素之间交互作用的多因素设计（常见的是析因设计和重复测量设计）的定量资料,对科学评价全部因素对观测结果的影响情况有着十分重要的意义。     

单组设计一元定性资料的统计推断与实验设计

In the former two issues of this periodical, we have introduced how to correctly estimate a suitable sample size and test power for quantitative data of a single-group design. This article will introduce the interval estimation and hypothesis testing of qualitative data of a single-group design by SAS.     

Triple-type Theory of Statistics and Its Role of Guidance for Scientific Research Work

Objective To offer a practical solution to the situation that many researchers make frequent errors and can hardly capture the essence of statistics despite various learning and training experiences in related courses, in the hope of improving the validity of statistical research design and statistical analysis, thereby improving the overall quality of world＇s research projects and academic publications to guarantee scientific reliability and exactness of our research endeavors. Methods Based on the experience in long-time teaching and research in statistics, the author points out the crux of misuse of statistics and proposes a ＂triple-type theory of statistics＂, which may promise a key to solving this issue. The ＂three-type theory＂ is summarized as follows： Any statistical problem in practical research presents itself in three types- expressive type, prototype and the standardized type. It is of critical importance to understand their correlations in a specific study before applying statistical analysis methods, which helps researchers ＂see the essence through the superficial phenomena＂, i.e. apply statistics rationally and properly. Results Our investigations reveal that for certain problems, the three types are mutually identical; for some problems, the ＂prototype＂ is the ＂standardized type＂; however, for some others, the three types are distinct from each other. Particularly in some multifactor experimental studies, researchers have committed the ＂incomplete control error＂ in setting experimental groups, which leads to the non-existence of the ＂standardized type＂ corresponding to the ＂prototype＂. This difficult problem could be resolved by introducing the concept and method of ＂splitting groups＂. Conclusion Once the three types and their correlations for each specific question are clarified, a good design of research, proper presentation and interpretation of data, a sound selection and application of statistical method can be carried out naturally and easily. Triple-type theory of statistics can help researchers avoid committing statistical errors or at least reduce the rate of misuse and therefore enhance the quality and level of our biomedical research. It may also facilitate the compilation of new textbooks and reform of teaching methodology, providing guidelines for the application and development of biomedical statistics.     

如何表达与描述定量资料（四）

在医学论文中,如何正确表达与描述定量资料,对于正确选择统计学中的假设检验方法至关重要。当试验涉及两个、三个或多个因素,且这些因素之间依据专业知识和重复试验次数多少认为可以考察其各种关系时,该定量资料就属于可以考察因素之间交互作用的定量资料。根据试验设计类型和专业知识,运用恰当的方法表达和描述可以考察因素之间交互作用的多因素设计（常见的是多因素析因设计和含区组因素析因设计）的定量资料,对科学评价全部因素对观测结果的影响情况有着十分重要的意义。     

如何表达与描述定量资料（五）

在生物医学科研中,经常会涉及多因素试验研究问题,若科研经费充足,人力和时间都允许,选择多因素析因设计来安排试验是非常好的,因为此设计类型可以分析各因素不同水平对结果的影响,还可分析全部因素各级交互作用对结果的影响。然而,当前面所提到的条件不允许时,就需要选用正交设计。有时,选用正交设计仍感到试验次数过多,     

Saffron （Crocus sativus L.） and major depressive disorder： a meta-analysis of randomized clinical trials

Due to safety concerns and side effects of many antidepressant medications, herbal psychopharmacology research has increased, and herbal remedies are becoming increasingly popular as alternatives to prescribed medications for the treatment of major depressive disorder （MDD）. Of these, accumulating trials reveal positive effects of the spice saffron （Crocus sativus L.） for the treatment of depression. A comprehensive and statistical review of the clinical trials examining the effects of saffron for treatment of MDD is warranted. OBJECTIVE： The purpose of this study was to conduct a meta-analysis of published randomized controlled trials examining the effects of saffron supplementation on symptoms of depression among participants with MDD. SEARCH STRATEGY： We conducted electronic and non-electronic searches to identify all relevant randomized, double-blind controlled trials. Reference lists of all retrieved articles were searched for relevant studies. INCLUSION CRITERIA： The criteria for study selection included the following： （1） adults （aged 18 and older） with symptoms of depression, （2） randomized controlled trial, （3） effects of saffron supplementation on depressive symptoms examined, and （4） study had either a placebo control or antidepressant comparison group. DATA EXTRACTION AND ANALYSIS： Using random effects modeling procedures, we calculated weighted mean effect sizes separately for the saffron supplementation vs placebo control groups, and for the saffron supplementation vs antidepressant groups. The methodological quality of all studies was assessed using the Jadad score. The computer software Comprehensive Meta- analysis 2 was used to analyze the data. RESULTS： Based on our pre-specified criteria, five randomized controlled trials （n = 2 placebo controlled trials, n = 3 antidepressant controlled trials） were included in our review. A large effect size was found for saffron supplementation vs placebo control in treating depressive symptoms （M ES = 1.62, P 〈 0.001）, revealing that saffron supplementation significantly reduced depression symptoms compared to the placebo control. A null effect size was evidenced between saffron supplementation and the antidepressant groups （M ES = -0.15） indicating that both treatments were similarly effective in reducing depression symptoms. The mean Jadad score was 5 indicating high quality of trials. CONCLUSION： Findings from clinical trials conducted to date indicate that saffron supplementation can improve symptoms of depression in adults with MDD. Larger clinical trials, conducted by research teams outside of Iran, with long-term follow-ups are needed before firm conclusions can be made regarding saffron＇s efficacy and safety for treating depressive symptoms.