Selecting variables for logistic regression

After you have specified that you want to perform a logistic regression run and selected the LR type and a data configuration, you select the drug (predictor) and event (response) variables to use in the run. You also have the option to select one or more variables as covariates (additional predictors).

1.         On the Select Variables page, click one variable in the Event variable list and one variable in the Drug variable list.

Available variables are those that are defined as event variables or drug variables in the data configuration, and that do not have associated data transformations.

Note: To treat combination drugs as if the subject had taken separate drugs, select a variable, such as Single Ingredient, as the drug variable.

2.         You can also select one or more variables in the Additional covariates list. Available variables are those that are defined as stratification variables in the data configuration, and that do not have associated data transformations. (To identify a variable as a stratification variable, edit the variable in the data configuration.)

Typically, a logistic regression run includes one or more additional covariates as predictors. Otherwise, the logistic regression can be subject to confounding due to potentially highly associated covariates being left out. Recommended covariates for an AERS-based configuration include the AgeGroup4, Gender, and Subset_Year variables.

To select one variable, click its name. To select multiple variables as covariates:

  1. Select multiple non-contiguous variables by holding down the Ctrl key while clicking each variable.
  2. Select multiple contiguous variables by clicking a variable, holding down the Shift key, and clicking another variable. Variables between and including those variables are selected.
  3. To deselect a variable, hold down the Ctrl key while clicking the variable.

Note: AERS data includes the variables STANDARD_STRATA and STANDARD_STRATA_INIT. These variables are not available as covariates in logistic regression runs.

3.         To set up a custom term for this data mining run, check Define custom terms.

4.         Click Next.

The Select Events page appears.

5.         Continue by selecting event values. If you checked Define Custom Terms, the Define Custom Terms page appears first.