ABSTRACT
Las revisiones sistemáticas y los metaanálisis se han consolidado como una herramienta fundamental para la práctica clínica basada en la evidencia. Inicialmente, el metaanálisis fue propuesto como una técnica que podría mejorar la precisión y la potencia estadística de la investigación procedente de estudios individuales con pequeño tamaño muestral. Sin embargo, uno de sus principales inconvenientes es que suelen comparar no más de 2 intervenciones alternativas a la vez. Los «metaanálisis en red» utilizan técnicas novedosas de análisis que permiten incorporar la información procedente de comparaciones directas e indirectas a partir de una red de estudios que examina los efectos de diversos tratamientos de una manera más completa. Pese a sus potenciales limitaciones, su aplicación en epidemiología clínica podría ser potencialmente útil en situaciones en las que existen varios tratamientos que se han comparado frente a un comparador común. Además, estas técnicas pueden ser relevantes ante una pregunta clínica o de investigación cuando existen múltiples tratamientos que deben ser considerados, o cuando se dispone tanto de información directa como indirecta en el cuerpo de la evidencia.
Systematic reviews and meta-analyses have been established as fundamental tools for evidence-based clinical practice. Initially, meta-analysis was proposed as a technique that could improve the precision and statistical power of individual studies research, with small sample sizes. However, one of its main drawbacks was related to usually comparing only 2 alternatives at a time. "Network metaanalyses" uses novel analytical techniques that allow information from direct and indirect comparisons to be incorporated from a network of studies that examine the effects of various treatments in a more comprehensive way. Despite potential limitations, its application in clinical epidemiology would most likely be useful in situations where there are several treatments that need to be compared against a common comparator. In addition, these techniques may be relevant to answer research questions that involve multiple treatments, or when both direct and indirect information are available in the body of evidence.
ABSTRACT
The purpose of the paper was to introduce the multiple comparison method among multiple means and the SAS implementation. The multiple comparison approaches could be subdivided into the pairwise comparisons， the comparisons of all treatment groups with a control group， the comparisons of the mean of each treatment group with the average of all groups， the approximate and simulation-based approach， the multi-stage testing and Bayesian method. Except for the Bayesian approach， the difference between other multiple comparison methods lied in the types of error that were controlled. Error types could be roughly divided into the following three categories， the comparisonwise error rate， the experimentwise error rate and the maximum experimentwise error rate. The multiple comparison methods constructed based on the control of different error rates were not all the same in the strength of inference to draw conclusions. This paper used the SAS software to analyze the examples and explained the output results.
ABSTRACT
The purpose of this paper was to introduce the prerequisites， basic ideas， calculation formulas and the SAS implementation of a single-factor multi-level design quantitative data univariate analysis of variance. The prerequisites included the independence， normality and homogeneity of variance. The core of the basic idea was the decomposition of the sum of squares of the total deviations for the mean. The test statistic F was constructed through the between-group mean square divided by the within-group （or called error） mean square. The result of analysis of variance was a general evaluation of the difference among all means of a factor with the whole levels. When it was found that the difference among all means of the factor was statistically significant， a specific approach needed to be adopted for the multiple comparisons about the multiple means of the factor. With the help of the SAS software， the paper performed the analysis of variances for two examples， and used three approaches to make the multiple comparisons among all means of a factor in one of the examples.
ABSTRACT
The purpose of the paper was to introduce the three special tests of the survival data and the SAS implementation. Specifically， it was the multiple comparisons， the trend test and the covariate test of the survival data. The multiple comparisons involved two situations： "the pairwise comparison" and "the comparison with control group". In the trend test， it involved two algorithms： "the log-rank test" and "the Wilcoxon test". In the covariate test， it involved "the single covariate test method" and "the multi-covariate test method of adding one covariate step by step". With the help of the SAS software and based on an example， this article implemented the three special tests mentioned above， explained the output results， and made statistical and professional conclusions.
ABSTRACT
Objective To explore the effectiveness of different multiple comparisons correction methods by comparing the detection rate and false positive rate of brain activation analysis using functional magnetic resonance imaging ( fMRI) data. Methods On the basis of task-based fMRI dataset ( including low-intensity and high-intensity stimuli condition,n=20) and resting-state fMRI dataset( n=32),brain acti-vation results were corrected by multiple comparsion correction methods in SPM and SnPM13 software,and the activation detection rate and false positive rate were compared with different correction methods. Results Voxel-or peak-based correction methods had relatively low false positive rate. When P<0. 05 after correction,the proportion of the subjects with false-positive were 0. 19 and 0. 16,and the number of false-pos-itive voxels were 404 and 2 448,respectively. But the two methods had low detection rate,which were more suitable for detecting strong activation. While cluster-based correction methods had relative high detection rate and high false positive rate. When P<0. 05 after correction,the proportion of the subjects with false-posi-tive were 0. 34 and 0. 38,and the number of false-positive voxels were 7 870 and 8 320,respectively. And thus they were more suitable for detecting weak activation. Group-level analysis could effectively reduce false positive rate. Conclusion In practice,researchers should choose a suitable correction method based on their specific research objectives and data to achieve a balance between the detection rate and false positive rate.
ABSTRACT
Objective@#To explore the effectiveness of different multiple comparisons correction methods by comparing the detection rate and false positive rate of brain activation analysis using functional magnetic resonance imaging (fMRI) data.@*Methods@#On the basis of task-based fMRI dataset (including low-intensity and high-intensity stimuli condition, n=20) and resting-state fMRI dataset(n=32), brain activation results were corrected by multiple comparsion correction methods in SPM and SnPM13 software, and the activation detection rate and false positive rate were compared with different correction methods.@*Results@#Voxel-or peak-based correction methods had relatively low false positive rate.When P<0.05 after correction, the proportion of the subjects with false-positive were 0.19 and 0.16, and the number of false-positive voxels were 404 and 2 448, respectively.But the two methods had low detection rate, which were more suitable for detecting strong activation.While cluster-based correction methods had relative high detection rate and high false positive rate.When P<0.05 after correction, the proportion of the subjects with false-positive were 0.34 and 0.38, and the number of false-positive voxels were 7 870 and 8 320, respectively.And thus they were more suitable for detecting weak activation. Group-level analysis could effectively reduce false positive rate.@*Conclusion@#In practice, researchers should choose a suitable correction method based on their specific research objectives and data to achieve a balance between the detection rate and false positive rate.
ABSTRACT
Objective We introduces a method MCP-Mod ( Multiple Comparisons and Modeling) ,which can estimate the dose of dose-response studies. Based on the application conditions and analysis method of MCP-Mod,we explore the power to detect a dose-response relationship and the accuracy of model selection and target dose estimation using simulation data. Meth-ods Firstly,the research introduces the fundamental theory and application condition of the MCP-Mod method. Then,simulate data is set to evaluate the power to detect a dose-response relationship and the accuracy of model selection and target dose esti-mation. Results In the simulation,in the aspect of the power to detect a dose-response relationship:6 groups of constant model have an approximately power 0f 0. 05 to detect the dose response relationship,which was similar toα=0. 05. For other parameter models,the larger the sample size,the higher power to detect a dose response relationship. In the aspect of accuracy of model selection:the larger the sample size,the higher degree to identify various models. In the candidate models,exponential model and quadratic model have a high degree of identification,but linear model and logarithmic function have a low degree of identifica-tion. In terms of the accuracy of the dose estimation:M
ABSTRACT
Objective We introduces a method MCP-Mod ( Multiple Comparisons and Modeling) ,which can estimate the dose of dose-response studies. Based on the application conditions and analysis method of MCP-Mod,we explore the power to detect a dose-response relationship and the accuracy of model selection and target dose estimation using simulation data. Meth-ods Firstly,the research introduces the fundamental theory and application condition of the MCP-Mod method. Then,simulate data is set to evaluate the power to detect a dose-response relationship and the accuracy of model selection and target dose esti-mation. Results In the simulation,in the aspect of the power to detect a dose-response relationship:6 groups of constant model have an approximately power 0f 0. 05 to detect the dose response relationship,which was similar toα=0. 05. For other parameter models,the larger the sample size,the higher power to detect a dose response relationship. In the aspect of accuracy of model selection:the larger the sample size,the higher degree to identify various models. In the candidate models,exponential model and quadratic model have a high degree of identification,but linear model and logarithmic function have a low degree of identifica-tion. In terms of the accuracy of the dose estimation:M
ABSTRACT
[Objective] Discussion about feeding Liuwei dihuang pil to different ages of mental retardation intel igence-type mice, then test their intel igence level and mechanism. [Methods]The changes of ethology expressions are determined in healthy KunMing mice of five groups divided randomly:male mice (n=5), female mice(n=20). Group A:normal control group;Group B:the group with innate kidney deficiency and lead ingested. Group C:the group of lead poisoning, and the offspring mice are administered pil s. Group D:the group with innate kidney deficiency and lead ingested, and the adult mice are adminis-tered pil s. Group E:the group with innate kidney deficiency and lead ingested, and both the adult mice and offspring mice are administered pil s. According to the proportion of male and female mice, make them mate, remove the male mice. Before delivery, do threat means and give 0.1%Lead acetate trihydrate solution to Groups B,C,D,E.Groups D,E have tonifying the kidney. Give routine food to Groups A,B,C. Female mice wil labor in 19~22 days after mating, getting the offspring mice. The offspring mice in different periods of growth are carried out ethology determination and analyzed.[Results]Newborn rat kid-ney stand test shows rat offspring is not better than the success rate of kidney emancipated group, juvenile rat kidney tests show space group behavior test per-formance superior to that of kidney group, and long-term kidney group(offspring, mother on behalf of both rat kidney) or female offspring than only on be-half of kidney groups, multiple comparisons between groups(LSD-t statistical analysis) are statistical y significant, A group compared with B group, C group, D groups respectively(t=27.50,8.10,9.30,3.00,P<0.05), B group compared with C group, D group, E group(t=19.40,18.20,27.80, P al <0.05), C group and D group have difference statistical y significant(t=8.40, P<0.05), D group and E group difference is statistical y significant(t=9.60, P<0.05), young rats tightrope test, kidney group to some extent, is stil better than not Bushen group, multiple comparisons between groups find that:A group and D group, B group and C group, B group and E group, C group and D group, D group and E group have differences statistical y significant(t=22.01, 25.41,20.90,32.11,27.60, P<0.05). [Conclusion]Liuwei Dihuang Pil can improve the intel igence level of mice to a certain extent.
ABSTRACT
As a nonparametric method,the Kruskal-Wallis test is widely used to compare three or more independent groups when an ordinal or interval level of data is available,especially when the assumptions of analysis of variance (ANOVA) are not met.If the Kruskal-Wallis statistic is statistically significant,Nemenyi test is an alternative method for further pairwise multiple comparisons to locate the source of significance.Unfortunately,most popular statistical packages do not integrate the Nemenyi test,which is not easy to be calculated by hand.We described the theory and applications of the Kruskal-Wallis and Nemenyi tests,and presented a flexible SAS macro to implement the two tests.The SAS macro was demonstrated by two examples from our cohort study in occupational epidemiology.It provides a useful tool for SAS users to test the differences among three or more independent groups using a nonparametric method.
ABSTRACT
Sisvar is a statistical analysis system, first released in 1996 although its development began in 1994. The first version was done in the programming language Pascal and compiled with Borland Turbo Pascal 3. Sisvar was developed to achieve some specific goals. The first objective was to obtain software that could be used directly on the statistical experimental course of the Department of Exact Science at the Federal University of Lavras. The second objective was to initiate the development of a genuinely Brazilian free software program that met the demands and peculiarities of research conducted in the country. The third goal was to present statistical analysis software for the Brazilian scientific community that would allow research results to be analyzed efficiently and reliably. All of the initial goals were achieved. Sisvar gained acceptance by the scientific community because it provides reliable, accurate, precise, simple and robust results, and allows users a greater degree of interactivity.
O Sisvar é um sistema de análise estatística que foi lançado em 1996, embora o seu desenvolvimento tenha sido iniciado em 1994. A primeira versão foi desenvolvida em linguagem de programação Pascal e compilada com o Borland Turbo Pascal 3. O Sisvar foi desenvolvido em virtude de algumas razões específicas. O primeiro objetivo foi o de obter um software que pudesse ser usado diretamente no curso de estatística experimental do Departamento de Ciências Exatas da Universidade Federal de Lavras. O segundo objetivo foi o de iniciar o desenvolvimento de um software genuinamente brasileiro, gratuito que atendesse às demandas e peculiaridades das pesquisas realizadas no país. O terceiro objetivo foi o de apresentar um software de análise estatística para a comunidade científica brasileira que permitisse que os resultados da pesquisa pudessem ser analisados de forma eficiente e confiável. Todos os objetivos iniciais foram atingidos. O motivo da aceitação Sisvar pela comunidade científica é decorrente do fato de que ele é capaz de permitir uma maior interatividade com o usuário e produzir análises confiáveis, pelo fato de elas serem exatas, precisas, simples e robustas.