 |
ClinStReport Applications |
||
|---|---|---|---|
|
SAS macros, DM system for clinical data management, analysis and reporting |
|||
| Technique Details | |||
|
1. The UNIVARIATE, MEAN, FREQ, GLM, NPAR1WAY, MIXED and LIFETEST procedures are used in the macro group to generate statistic parameters 2. All the variables in the analysis data can be automatically located and analyzed, however the macro parameter vKeep or vDrop can be assigned to choose the right variables to analysis, the usage of vKeep and vDrop is same as SAS KEEP & DROP statement. 3. The type of variables (continuous or categorical) can be auto-identified according to their formats, therefore before runningl this macro, the variables in the analyzed data set should be correctly formatted and labeled. 4. For continuous variable, the decimal place of MEAN will be one digit less than that of SD, and one more digit than the raw data, which format will be automatically detected. 5. Methods used to calculate p value should be assigned correctly if requested. This is not automatic chosen for occasionally normal distributed data can not be detected when sample size is much small. The embedded methods for analysis of continuous data is ANOVA (PROC GLM) and NPAR1WAY (Wilcoxon/ Kruskal-Wallis test), for categorical data is Chi Square test, Fisher’s Exact Test and CMH Ridit. 6. Handling missing values in categorical data analysis and reporting: missing data will be excluded while performing Chi square test or McNemar/kappa test except explicit code is given for missing response(e.g. 1=YES, 2=NO, 9=Unknown). A parameter missing is used to control how to display missing response in summary tables. |
|||
|
|||
| More... |
|
||
|
Addr: Room 203, No 8, 165 Mudan Road, Pudong, Shanghai, P.R. China 201204 Tel: (510)-331-2327(US) ; + 86 13916348130(CN) Copyright© Shanghai ClinStReport Co., Ltd. All rights reserved. |
|||