It is often difficult to conclude that improvements in survival with time are due to a screening programme alone. Although a reduction in the death rate from a given cancer may reflect the benefits of early detection or improved treatment, the benefits may also result from lead time bias and over-diagnosis, the former resulting in longer survival of screen-identified cancers because the time before the cancer would have been clinically diagnosed is included in calculations. Furthermore, recent reviews on randomised clinical trials of cancer screening have provided strong evidence that misclassifications in causes of death have been a major problem, leading to an over-estimation of the effectiveness (or alternatively an under-estimation of potential harm) of screening.