ABSTRACT
Survival refers to the period of time from cancer diagnosis to death. Lead-time bias refers to the interval between diagnosis of cancer by screening and usual clinical detection. Leadtime bias may make it appear that screening prolongs survival, when, in fact, it simply extends the period of time over which the disease is observed. As screening advances the time of breast cancer diagnosis, patients with screen-detected cancers will inevitably have better survival rates than those with clinically detected cancers, even if screening does nothing to delay the time of death. Thus, retrospective studies comparing survival between screened and unscreened populations fail to account for lead-time bias, and are fl awed.
