Multiple regressions

Learn me by follow me exercises

This module tests for associations between risk factors and outcomes with different adjusting variables. It can quantify the exposure effect with or without other covariates.

Given a set of outcomes, risk factors, and covariates, this module runs regressions for each outcome with each or all risk factors.

The risk factor and outcome could be continuous or categorical variables.  If the risk factors are categorical variables only, you can direct Empower(S) to list Mean+SD (if outcome is continuous) or N (%) (if outcome is binary) for the outcomes among each subgroup of each risk factor.

Stratified variables could be added to direct the module to do the regressions for each group as well as for the whole study population.

Up to 3 different adjusting models, and up to 2 levels of stratifications could be specified.

This module provides flexibility for designing the output table format (rows and columns). You can specify the columns by stratified subgroups, or outcomes, or risk factors or models; you can order the rows by stratified subgroups, outcomes, risk factors or models.

The module automatically detects whether the outcome variable is binary or continuous and selects logistic or linear regression as appropriate.  You also can specify the distribution and link function for each outcome manually, right click the variable and then select “Change functions”. In the popup window, select the distribution and link function.

What it’s used for

Up to 3 models with different set of adjustment covariates could be specified, which helps to get more insight into how the relationship of risk factor(s) and outcome will be affected by different adjustments.

Screen shot of sample designs:

t4_input.gif    t4_input2.gif    t4_input3.gif

Sample output tables:

t4_mreg.gif    t4_mreg_2.gif    t4_mreg_3.gif