Researchers are trying to assess the effectiveness of a new blood pressure medication. Using their data, they calculate a simple linear regression model that predicts systolic blood pressure (SBP) in terms of BP Meds (where 0 means the new medication is given and 1 means a placebo is give). The results are shown in the last row of the middle 2 columns of the table below. The researchers believe that Age, Gender, and BMI might be confounders. They calculate simple linear models for each of these variables as shown in the table where SBP is the response variable in each model. Then they calculate a multiple regression model that predicts SBP in terms of all 4 variables. The results are given in the last 2 columns on the table. Based on these results, is the association between BP meds and SBP confounded by Age, Gender or BMI? Provide a brief (1-2 sentences) explanation.
Simple Models | Multiple Regression | |||
b | p | b | p | |
Age | 1.03 | <.0001 | 0.86 | <.0001 |
Male | -2.26 | 0.0009 | -2.22 | 0.0002 |
BMI | 1.8 | <.0001 | 1.48 | <.0001 |
BP Meds | 33.38 | <.0001 | 24.12 | <.0001 |
From the given summary it seems that the association between BP and SBP are confounded by Age, Gender or BMI,
But after looking the results of the simple regression and multiple regression it can be said that main coufounding variable is AGE and due to age the BP changes and that confounds or cause the association between BP and SBP.
Also, if you observe the Gender values the they arent differe much, hence Mostly AGE is the causing the association or confounding the association between the variables.
Get Answers For Free
Most questions answered within 1 hours.