Is there a difference in significant contributing factors to a prediction model for patients who are in the control group? If so, what are the factors and their p-values in the final model?