《临床统计学介绍39市公开课一等奖省优质课赛课一等奖课件.pptx》由会员分享，可在线阅读，更多相关《临床统计学介绍39市公开课一等奖省优质课赛课一等奖课件.pptx（41页珍藏版）》请在taowenge.com淘文阁网|工程机械CAD图纸|机械工程制图|CAD装配图下载|SolidWorks_CaTia_CAD_UG_PROE_设计图分享下载上搜索。
1、为何要做统计分析？统计分析目标是应用样本样本 资料信息,作出相关研研究总体究总体有效推测。应用概要性指标概要性指标描述样本资料来实现。这些概要性指标保留了足够信息足够信息 去预计研究总体特征。61第1页关于总体临床研究问题关于总体临床研究问题在发展中国家，人工喂养相比母乳喂养能否增加母亲为HIV阳性婴儿生存率？怎样建立一个心脏搭桥手术后生存率模型？病人特征能否预测术后生存率？相比内科治疗，搭桥手术后1，3，5年生存率能否改进？局部治疗小肝癌能否代替外科手术切除？根治术后应用大剂量干扰素能否降低肝癌复发率？62第2页今天主题 n总体，样本和个体n资料类型：Continuous vs.catego
2、ricaln怎样描述资料?统计量 和图n测量集中趋势和离散趋势n标准误和95%可信区间n依据数据选择适当统计方法n诊疗试验评价63第3页总体，样本和个体“Aristotle maintained that women have fewer teeth than men;although he was twice married,it never occurred to him to verify this statement by examining his wives mouths.”-Sir Bertrand Russell,The Impact of Science on Society
3、,1952.“It is a capital mistake to theorize before you have data.”-Sir Arthur Conan Doyle,Scandal in Bohemia.64第4页总体，样本和个体And,for another viewpoint:“If your experiment needs statistics,you ought to have done a better experiment.”Ernest Rutherford.The bench science perspective:you can control all the
4、variables!Clinicians,however,know better human variation is large,and often inexplicable.Statistics help us describe it and generalize at least enough to improve our ability to practice medicine.65第5页总体，样本和个体Aristotle 推测了一个女性总体女性总体(比较男性总体).他实际上手头就有一个包含2个女人样本个女人样本，他能对这个样本中2个个体个体进行数牙。The population is
5、 the collection of all people about whom you would like to ask a research question.This might be a fairly clear-cut easily defined set of people:“What proportion of people 65 or older in the US today have Alzheimers disease?”Or it might be a more hypothetical group:“How much of a reduction in sympto
6、matic days could a person expect if treated with a new antiviral for flu?”66第6页总体，样本和个体实际上，我们不可能去研究总体中每一个对象。所以，我们研究一个样本样本,并将其推广到整个人群。样本样本量量 是样本中个体个体 数目(而不是对每个研究对象测量指标数目!)好研究设计能帮助我们得到一个 代表性好样本。好统计分析能帮助我们取得关于总体问题答案。67第7页例子：HCC裸鼠转移模型免疫重建对照组CD331.5%14.2%CD4 XX XXCD8 XX XX*2个水平：裸鼠 细胞68第8页今天主题 n总体，样本和个体n资
7、料类型：Continuous vs.categoricaln怎样描述资料?统计量 和图n测量集中趋势和离散趋势n标准误和95%可信区间n依据数据选择适当统计方法n诊疗试验评价69第9页数据类型n计量资料 Quantitative:“how much?”连续变量连续变量:年纪,体重,身高,血压 实际数值实际数值:家庭儿女数,住院天数n分类资料 Categorical:“what type?”等级变量:肿瘤分期(I,II,III);好 中 差名义变量：男/女;健康/生病;ABO血型610第10页数据类型数据类型转换n计量数据可转换成份类数据：normal(value)vs.abnormal;“yo
8、ung,middle-aged,old”n将连续变量转换成等级变量降低了资料信息量，从而造成统计学检验敏感度或把握度下降611第11页今天主题 n总体，样本和个体n资料类型：Continuous vs.categoricaln怎样描述资料?统计量 和图n测量集中趋势和离散趋势n标准误和95%可信区间n依据数据选择适当统计方法n诊疗试验评价612第12页Notes:vertical axis can be count or percentin the above example,counts do not add to 74 individuals can have multiple risk
9、factorstabular presentation may be more parsimonious for such dataN=74分类资料统计描述计数百分比613第13页分类数据统计描述n组成比n率n百分比 vs 率n标化614第14页下面是一组年纪数据（11例）21,32,34,34,42,44,46,48,52,56,64年纪是一个计量变量，所以假如用条图就不适当。我们更感兴趣是年纪分布一些特征：年纪分别中心点在哪里？如平均数年纪变异又是怎样?是不是有些数据跟绝大部分数据差得很多（outliers）借助视觉工具帮助我们回答这些问题.定量数据统计描述定量数据统计描述615第15页计
10、量数据统计描述计量数据统计描述图表1.Stem and Leaf plot2.Histogram3.Boxplot数字1.Location-mean,median,mode.2.Spread-range,variance,standard deviation，percentile3.Shape-skewness*例外：生存资料描述616第16页We could group the data and tally the frequencies:But why“hide”the details?Instead,well use the 10s place as stems and the unit
11、s as leaves:20:X30:XXX40:XXXX50:XX60:X2*|13*|2444*|24685*|266*|4Stem and Leaf Diagramstem&leaf plotFor small datasets617第17页Examples平均数方差中位数百分位数outlier618第18页今天主题 n总体，样本和个体n资料类型：Continuous vs.categoricaln怎样描述资料?统计量 和图n测量集中趋势和离散趋势n标准误和95%可信区间n依据数据选择适当统计方法n诊疗试验评价619第19页集中趋势n算术平均数：n几何平均数n中位数620第20页平均数和
12、中位数比较平均数和中位数比较nMean is sensitive to a few very large(or small)values-“outliers”nMedian is“resistant”to outliersnMean is attractive mathematicallyn50%of sample is above the median,50%of sample is below the median.621第21页离散趋势Variation is important!622第22页离散趋势n方差n标准差n百分位数：IQR=Q.75-Q.25 623第23页今天主题 n总体，
13、样本和个体n资料类型：Continuous vs.categoricaln怎样描述资料?统计量 和图n测量集中趋势和离散趋势n标准误和95%可信区间n依据数据选择适当统计方法n诊疗试验评价624第24页标准误和95%可信区间n描述样本：平均数，标准差n？总体：n为了预计总体平均数，需要计算标准误n标准误标准差/样本量n总体均数95CI:n样本平均数1.96*标准误 论文中惯用625第25页标准差 vs均数标准误(when do you use one,but not the other?)n标准差标准差用于描述：量化样本均数周围变异.当确定两个样本是否来自于同一总体时，标准差是一个主要统计量。
14、nCentral limit theorem;“同一总体中样本均数呈正态分布”n样本均数标准误标准误用于样本均数预计总体均数。标准误是一个主要统计量，用于计算样本均数可信度，取决于标准差和样本量。但实际上二者并不独立，当样本量增加时，标准差往往降低。626第26页正态分布(basis of statistical inference for many populations )Mean=median=mode.all=same value in the distribution remember:68.3%of data is between -1.00 s.d.and +1.00 s.d.9
15、5.0%“-1.96 s.d.and +1.96 s.d.95.5%“-2.00 s.d.and +2.00 s.d.99.7%“-3.00 s.d.and +3.00 s.d.627第27页今天主题 n总体，样本和个体n资料类型：Continuous vs.categoricaln怎样描述资料?统计量 和图n测量集中趋势和离散趋势n标准误和95%可信区间n依据数据选择适当统计方法n诊疗试验评价628第28页推断性统计推断性统计推广结论:样本总体评价证据强度比较预测629第29页计量资料统计方法正态分布非正态分布配对资料（配对资料（2组组)配对t检验符号检验符号等级检验成组比较成组比较 (2组
16、）组）成组比较t检验Wilcoxon Mann&Whitney中位数检验配伍组比较配伍组比较随机区组方差分析非参数配伍组比较M检验多组比较多组比较完全随机设计方差分析非参数多组比较H检验630第30页列联表分析行名义变量等级变量名义变量n普通联络：nPearsons 2行平均得分：（趋势分析）等级变量行平均得分：2 （趋势分析）相关分析：cmh：2 列*四格表是全一致631第31页Make predictions:回归分析n应变量：n普通定量变量 线性分析n等级或名义变量Logistic 回归n时间变量 Cox回归632第32页Descriptive epidemiology:pattern
17、of occurrencePrevalence of HIV+and community Mosquito indexr =.83 r-squared=.92 *p .001 p .001 *201510 5 00 2 4 6 8 10 12 14 16 18 20 22Index of community mosquito infestation HIV+633第33页今天主题 n总体，样本和个体n资料类型：Continuous vs.categoricaln怎样描述资料?统计量 和图n测量集中趋势和离散趋势n标准误和95%可信区间n依据数据选择适当统计方法n诊疗试验评价634第34页诊疗试
18、验评价n试验设计635第35页诊疗试验设计636第36页诊疗试验评价金标准有病金标准无病试验ab试验cd敏感度a/a+c特异度d/b+d阳性预测值a/a+b阴性预测值d/c+d阳性拟然比敏感度/1特异度阴性拟然比1敏感度/特异度637第37页医学论文中通常报道哪些?大多数研究报道平均数（正态）或中位数（非正态）有些研究报道标准差和/或标准误。Be careful!有时会看到图中有一个error bar，could be either.假如资料非正态(偏态，多峰，尾巴很长或很短等）,往往报道中位数和百分位数，而不是均数和标准差.写文章时一定有根根本研究所要回答问题：Do you want to
19、ask about the average or typical person?Or do you want to figure out how unusual your patient might be?638第38页通常流行病学（科学）路径n1.确定一个问题问题:clinical suspicion;case series;review of medical literaturen2.组织一个假设假设 (asking the right question);good hypotheses are:Specific,Measurable,and Plausiblen3.检验假设检验假设 (a
20、ssumptions vs.type of data)n4.再验证验证 always Question the VALIDITY of the result(s):Chance;Bias;and Causality 639第39页结论准确性nChance:role of random error in outcome measure(s)(p-value;power of the study and the confidence interval)-largely determined by sample sizenBias:role of systematic error in outcom
21、e measure(s)nSelection bias -subjects not representativnInformation bias -error(s)in subject data/classificationnConfounding -3rd variable(causal)assoc.w/both X and Y640第40页oZr%u(x+B2E5H9KcOfRiUmXp#s&v)z0C3F7IaLdPgSjVnYq$t*w-A1D4G8JbNeQhTlWoZr%u(y+B2E6H9KcOfRjUmXp!s&v)z0C4F7IaMdPgSkVnZq$t*x-A1D5G8Kb
22、NeQiTlWo#r%v(y+B3E6H9LcOfRjUmYp!s&w)z0C4F7JaMdPhSkVnZq$u*x-A2D5G8KbNfQiTlXo#r%v(y0B3E6I9LcOgRjVmYp!t&w)z1C4G7JaMePhSkWnZq$u*x+A2D5H8KbNfQiUlXo#s%v(y0B3F6I9LdOgRjVmYq!t&w-z1C4G7JbMePhTkWnZr$u(x+A2E5H8KcNfRiUlXp#s%v)y0B3F6IaLdOgSjVmYq!t*w-z1D4G7JbMeQhTkWoZr$u(x+B2E5H9KcNfRiUmXp#s&v)y0C3F7IaLdPgSjVnYq$
23、t*w-A1D4G8JbNeQhTlWoZr%u(x+B2E6H9KcOfRiUmXp!s&v)z0C3F7IaMdPgSkVnYq$t*x-A1D5G8JbNeQiTlWo#r%u(y+B3E6H9LcOfRjUmYp!s&w)z0C4F7IaMdPhSkVnZq$t*x-A2D5G8KbNeQiTlXo#r%v(y+B3E6I9LcOgRjUmYp!t&w)z1C4F7JaMePhSkWnZq$u*x+A2D5H8KbNfQiUlXo#s%v(y0B3E6I9LdOgRjVmYp!t&w-z1C4G7JaMePhTkWnZr$u*x+A2E5H8KcNfQiUlXlXp#s%v)y0B3F
24、6IaLdOgSjVmYq!t*w-z1D4G7JbMeQhTkWoZr$u(x+A2E5H9KcNfRiUlXp#s&v)y0C3F6IaLdPgSjVnYq!t*w-A1D4G8JbMeQhTlWoZr%u(x+B2E6H9KcOfRiUmXp!s&v)z0C3F7IaMdPgSkVnYq$t*w-A1D5G8JbNeQhTlWo#r%u(y+B2E6H9LcOfRjUmXp!s&w)z0C4F7IaMdPhSkVnZq$t*x-A2D5G8KbNeQiTlXo#r%v(y+B3E6H9LcOgRjUmYp!s&w)z1C4F7JaMdPhSkWnZq$u*x-A2D5H8KbNfQiTl
25、Xo#s%v(y0B3E6I9LdOgRjVmYp!t&w-z1C4G7JaMePhSkWnZr$u*x+A2D5H8KcNfQiUlXo#s%v)y0B3F6I9LdOgSjVmYq!t&w-z1D4G7JbMePhTkWoZr$u(x+A2E5H9KcNfRiUlXp#s&v)y0C3F6IaLdOgSjVnYq!t*w-z1D4G8JbMeQhTkWoZr%u(x+B2E5H9KcOfRiUmXp#s&v)z0C3F7IaLdPgSkVnYq$t*w-A1D5G8JbNeQhTlWoZr%u(y+B2E6H9KcOfRjUmXp!s&v)z0C4F7IaMdPgSkVnZq$t*x-A1
26、D5G8KbNeQiTlWo#r%v(y+B3E6H9LcOgRjUmYp!s&w)z1C4F7JaMdPhSkVnZq$u*x-A2D5G8KbNfQiTlXo#r%v(y0B3E6I9LcOgRjVmYp!t&w)z1C4G7JaMePhSkWnZr$u*x+A2D5H8KcNfQiUlXo#s%v(y0B3F6I9LdOgRjVmYq!t&w-z1C4G7JbMePhTkWnZr$u(x+A2E5H8KcNfRiUlXp#s%v)y0C3F6IaLdOgSjVnYq!t*w-z1D4G7JbMeQhTkWoZr$u(x+B2E5H9KcNfRiUmXp#s&v)y0C3F7IaLdPgS
27、jVnYq$t*w-A1D4G8JbNeQhTlWoZr%u(y+B2E6H9KcOfRjUmXp!s&v)z0C3F7F7IaMdPgSkVnYq$t*x-A1D5G8JbNeQiTlWo#r%u(y+B3E6H9LcOfRjUmYp!s&w)z0C4F7JaMdPhSkVnZq$u*x-A2D5G8KbNfQiTlXo#r%v(y+B3E6I9LcOgRjUmYp!t&w)z1C4F7JaMePhSkWnZq$u*x+A2D5H8KbNfQiUlXo#s%v(y0B3F6I9LdOgRjVmYq!t&w-z1C4G7JbMePhTkWnZr$u*x+A2E5H8KcNfQiUlXp#s%v
28、)y0B3F6IaLdOgSjVmYq!t*w-z1D4G7JbMeQhTkWoZr$u(x+B2E5H9KcNfRiUmXp#s&v)y0C3F6IaLdPgSjVnYq!t*w-A1D4G8JbMeQhTlWoZr%u(x+B2E6H9KcOfRiUmXp!s&v)z0C3F7IaMdPgSkVnYq$t*x-A1D5G8JbNeQiTlWo#r%u(y+B2E6H9LcOfRjUmXp!s&w)z0C4F7IaMdPhSkVnZq$t*x-A2D5G8KbNeQiTlXo#r%v(y+B3E6I9LcOgRjUmYp!t&w)z1C4F7JaMdPhSkWnZq$u*x-A2D5H8Kb
29、NfQiTlXo#s%v(y0B3E6I9LdOgRjVmYp!t&w-z1C4G7JaMePhTkWnZr$u*x+A2E5H8KcNfQiUlXo#s%v)y0B3F6I9LdOgSjVmYq!t&w-z1D4G7JbMePhTkWoZr$u(x+A2E5H9KcNfRiUlXp#s&v)y0C3F6IaLdPgSjVnYq!t*w-A1D4G8JbMeQhTkWoZr%u(x+B2E5H9KcOfRiUmXp#s&v)z0C3F7IaLdPgSkVnYq$t*w*w-A1D5G8JbNeQhTlWo#r%u(y+B2E6H9LcOfRjUmXp!s&w)z0C4F7IaMdPhSkVnZ
30、q$t*x-A1D5G8KbNeQiTlWo#r%v(y+B3E6H9LcOgRjUmYp!s&w)z1C4F7JaMdPhSkWnZq$u*x-A2D5H8KbNfQiTlXo#s%v(y0B3E6I9LcOgRjVmYp!t&w)z1C4G7JaMePhSkWnZr$u*x+A2D5H8KcNfQiUlXo#s%v)y0B3F6I9LdOgSjVmYq!t&w-z1D4G7JbMePhTkWnZr$u(x+A2E5H8KcNfRiUlXp#s%v)y0C3F6IaLdOgSjVnYq!t*w-z1D4G8JbMeQhTkWoZr%u(x+B2E5H9KcOfRiUmXp#s&v)z0C3F
31、7IaLdPgSjVnYq$t*w-A1D4G8JbNeQhTlWoZr%u(y+B2E6H9KcOfRjUmXp!s&v)z0C4F7IaMdPgSgSkVnZq$t*x-A1D5G8KbNeQiTlWo#r%v(y+B3E6H9LcOfRjUmYp!s&w)z0C4F7JaMdPhSkVnZq$u*x-A2D5G8KbNfQiTlXo#r%v(y0B3E6I9LcOgRjVmYp!t&w)z1C4G7JaMePhSkWnZr$u*x+A2D5H8KbNfQiUlXo#s%v(y0B3F6I9LdOgRjVmYq!t&w-z1C4G7JbMePhTkWnZr$u(x+A2E5H8KcNfRi
32、UlXp#s%v)y0C3F6IaLdOgSjVmYq!t*w-z1D4G7JbMeQhTkWoZr$u(x+B2E5H9KcNfRiUmXp#s&v)y0C3F7IaLdPgSjVnYq$t*w-A1D4G8JbNeQhTlWoZr%u(y+B2E6H9KcOfRiUmXp!s&v)z0C3F7IaMdPgSkVnYq$t*x-A1D5G8JbNeQiTlWo#r%u(y+B3E6H9LcOfRjUmYp!s&w)z0C4F7JaMdPhSkVnZq$t*x-A2D5G8KbNeQiTlXo#r%v(y+B3E6I9LcOgRjUmYp!t&w)z1C4F7JaMePhSkWnZq$u*x+
33、A2D5H8KbNfQiUlXo#s%v(y0B3E6I9LdOdOgRjVmYq!t&w-z1C4G7JaMePhTkWnZr$u*x+A2E5H8KcNfQiUlXp#s%v)y0B3F6IaLdOgSjVmYq!t*w-z1D4G7JbMeQhTkWoZr$u(x+B2E5H9KcNfRiUlXp#s&v)y0C3F6IaLdPgSjVnYq!t*w-A1D4G8JbMeQhTlWoZr%u(x+B2E6H9KcOfRiUmXp!s&v)z0C3F7IaMdPgSkVnYq$t*x-A1D5G8JbNeQhTlWo#r%u(y+B2E6H9LcOfRjUmXp!s&w)z0C4F7IaM
34、dPhSkVnZq$t*x-A2D5G8KbNeQiTlXo#r%v(y+B3E6I9LcOgRjUmYp!s&w)z1C4F7JaMdPhSkWnZq$u*x-A2D5H8KbNfQiTlXo#s%v(y0B3E6I9LdOgRjVmYp!t&w-z1C4G7JaMePhTkWnZr$u*x+A2D5H8KcNfQiUlXo#s%v)y0B3F6I9LdOgSjVmYq!t&w-z1D4G7JbMePhTkWoZr$u(x+A2E5H9KcNfRiUlXp#s&v)y0C3F6IaLdPgSjVnYq!t*w*w-A1D4G8JbMeQhTkWoZr%u(x+B2E5H9KcOfRiUmXp
35、#s&v)z0C3F7IaLdPgSkVnYq$t*w-A1D5G8JbNeQhTlWo#r%u(y+B2E6H9LcOfRjUmXp!s&v)z0C4F7IaMdPgSkVnZq$t*x-A1D5G8KbNeQiTlWo#r%v(y+B3E6H9LcOgRjUmYp!s&w)z1C4F7JaMdPhSkWnZq$u*x-A2D5H8KbNfQiTlXo#r%v(y0B3E6I9LcOgRjVmYp!t&w)z1C4G7JaMePhSkWnZr$u*x+A2D5H8KcNfQiUlXo#s%v)y0B3F6I9LdOgSjVmYq!t&w-z1C4G7JbMePePhTkWoZr$u(x+A2
36、E5H8KcNfRiUlXp#s%v)y0C3F6IaLdOgSjVnYq!t*w-z1D4G8JbMeQhTkWoZr%u(x+B2E5H9KcOfRiUmXp#s&v)z0C3F7IaLdPgSjVnYq$t*w-A1D4G8JbNeQhTlWoZr%u(y+B2E6H9KcOfRjUmXp!s&v)z0C4F7IaMdPgSkVnZq$t*x-A1D5G8KbNeQiTlWo#r%u(y+B3E6H9LcOfRjUmYp!s&w)z0C4F7JaMdPhSkVnZq$u*x-A2D5G8KbNfQiTlXo#r%v(y0B3E6I9LcOgRjVmYp!t&w)z1C4G7JaMePhS
37、kWnZq$u*x+A2D5H8KbNfQiUlXo#s%v(y0B3F6I9LdOgRjVmYq!t&w-z-z1C4G7JbMePhTkWnZr$u(x+A2E5H8KcNfRiUlXp#s%v)y0C3F6IaLdOgSjVnYq!t*w-z1D4G7JbMeQhTkWoZr$u(x+B2E5H9KcNfRiUmXp#s&v)y0C3F7IaLdPgSjVnYq$t*w-A1D4G8JbNeQhTlWoZr%u(y+B2E6H9KcOfRiUmXp!s&v)z0C3F7IaMdPgSkVnYq$t*x-A1D5G8JbNeQiTlWo#r%u(y+B3E6H9LcOfRjUmYp!s&w)z0C4F7JaMdPhSkVnZq$t*x-A2D5G8KbNeQiTlXo#r%v(y+B3E6I9LcOgRjUmYp!t&w)z1C4F641第41页