Bob run a logistic regression model to predict the risk of readmission within 30
ID: 3154619 • Letter: B
Question
Bob run a logistic regression model to predict the risk of readmission within 30 days based on a number of explanatory variables. Results presented indicate that the coefficient for the explanatory binary variable “Lymphocyte % is low” is estimated to be 0.0128. Odds ratio is defined as p/(1-p), where p is the probability of readmission within 30 days. Consider a patient with low lymphocyte count (Lymphocyte % is low = 1) with p=0.4, that is, with a 40% chance of readmission within 30 days. If the patient responds to treatment so that lymphocyte count is no longer low (Lymphocyte % is low = 0), what is the patient’s new probability pNew of readmission within 30 days. Assume that all the other explanatory variables remain unchanged.
Explanation / Answer
, what is the patient’s new probability pNew of readmission within 30 days. Assume that all the other explanatory variables remain unchanged.
Patients new probability p new of readmission within 30 days
0.0128 / ( 1 - 0.0128) = 0.0129
0.4 * 0.0129 = 0.005186