After fitting your model, you can determine the unmeasured confounder
needed to tip your analysis. This unmeasured confounder is determined by
two quantities, the relationship between the exposure and the unmeasured
confounder (if the unmeasured confounder is continuous, this is
indicated with exposure_confounder_effect, if binary, with
exposed_confounder_prev and unexposed_confounder_prev), and the
relationship between the unmeasured confounder and outcome
confounder_outcome_effect. Using this