ABSTRACT
The relationship between initial disease status and subsequent change following treatment has attracted great interest in medical and dental epidemiology. It is an amazing example of how a deceptively simple question has managed to confuse generations of researchers.
