![]() ![]() Probability and Odds of Developing CA in Each of the Treatment Arms Depending on the study design, one can compare either the probability of events or the odds of an event occurring. The answer to questions like this hinge on a comparison of the frequencies of these health events in the two treatment groups. We might be interested in knowing whether the frequency of coronary artery abnormalities differed between these two groups, i.e., was one of these associated with fewer abnormalities. One goal of the study was to compare the probabilities of developing CA abnormalities given treatment with Aspirin (ASA) or Gamma globulin (GG). The primary outcome variable in the Kawasaki trials was development of coronary artery (CA) abnormalities, a dichotomous variable. ![]() ![]() In general, 2 x 2 tables summarize the frequency of health-related (or other) events among different groups, as illustrated below in which Group 1 may represent patients who received a standard therapy, and Group 2 might be patients who received a new experimental therapy. To summarize the association between two categorical variables with R and C levels, we create cross-tabulations, or RxC tables ("Row"x"Column" or contingency tables), which summarize the observed frequencies of categorical outcomes among different groups of subjects. Categorical variables may represent the development of a disease, an increase of disease severity, mortality, or any other variable that consists of two or more levels. ![]()
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |