“

One has to be cautious when interpreting studies using medical records. They can be fraught with challenges in how diseases are coded and studied, any anytime you ask people to recall their behaviors, such as drinking, it leaves room for memory errors.