Inference in Case-Control Studies