Over-diagnosis plays havoc with our understanding of cancer statistics. Because over-diagnosis effectively changes a healthy person into a diseased one, it causes overestimations of the sensitivity, specificity, and positive predictive value of screening tests and the incidence of disease (13). As the MLP and a recent analysis of Surveillance, Epidemiology, and End Results (SEER)1 data illustrate (14), over-diagnosis also markedly increases the length of survival, regardless of whether screening or associated treatments are actually effective. However, over-diagnosis does not reduce disease-specific mortality because treating subjects with pseudo-disease does not help those who have real disease. Consequently, disease-specific mortality is the most valid end point for the evaluation of screening effectiveness.14