Document RpkkvoqNELw87nNedVq79v6vz

For discrete variables, the statistical analyses will employ simple stratified analyses using Mantel-Haenszel models (Mantel and Haenszel, 1959) or logistic regression models (Breslow and Day, 1980,1987) as appropriate. Statistical analyses will employ simple stratified analyses using Mantel-Haenszel statistics (Mantel and Haenszel, 1959), analysis of variance techniques and regression modeling. Both parametric and non-parametric techniques will be used, depending on required statistical distributional assumptions for parametric analyses. Statistical analyses will primarily be carried out in SAS, although other packages with common epidemiologic statistical routines (EGRET, EPICURE, STATA) will be used to supplement SAS analyses. Continuous measures will be analyzed by appropriate linear model analyses (e.g. analysis of variance, regression, analysis of covariance [Snedecor and Cochran, 1989]). For the case of repeated or serial continuous measures, such as serial WBC, the analysis will be based on mixedeffect models appropriate for repeated measures on a single subject over time (Milliken and Johnson, 1984). A special case for analysis is the role of genetic polymorphisms on exposure/response relationships that will be explored in subsequent analyses. For this assessment, each outcome will be characterized as a dichotomy and genetic polymorphisms, exposure etc. will be predictor variables (e.g. presence absence of chromosome 7q- deletions). In this manner, statistical analyses appropriate for case control studies can be employed as in Andrieu and Goldstein, (1998). References Andrieu N and Goldstein AM Epidemiologic and Genetic Approaches in the Study of GeneEnvironment Interaction: an Overview of Available Methods. Epidemiologic Reviews 1998 20:137-147. Breslow, NE and Day, NE, Statistical Methods in Cancer Research, Vol. I - The Analysis of Case-control studies. IARC, Lyon, France, 1980. Breslow, NE and Day, NE, Statistical Methods in Cancer Research, Vol.II - The Design and Analysis of Cohort studies. IARC, Lyon, France, 1987. Mantel, Nand Haenszel, W Statistical Aspects of the Analysis of Data from Retrospective studies of Disease. J Natl. Inst. 22: 719-748,1959. Milliken, GA and Johnson, DE Analysis of Messy Data, Vol I: Designed Experiments, Van Nostrand Reinhold, Co, NY, 1984. Snedecor, GW and Cochran, WG, Statistical Methods, 8th Ed. Iowa State Univ. Press, Ames, Iowa, 1989. 118 SH ELL-MCCLU RG-059563