Findinf confounders in a data

[Zahoor Khan]

I am working on a data set where I am trying to find association between hospitalization for heart attack - (myocardial infarction) - and pulse pressure. However, there could be several confounders as well. I am using logistic regression for the dichotomous outcome variable, using the continuous "pulse pressure - which is equal to systolic - diastolic blood pressure", . For the confounders in the data, should I use the logistic regression or linear regression to find and verify a confounder or confounders.

[Leonardo Guizzetti]

Confounding variables cannot be found by looking at the data. You need a causal diagram/model for how your variables relate to each other to determine those.