Resumen
An important component in the study of survival in breast cancer patients is how the tumor is detected. This involves examining whether and to what extent survival differs when the tumor is detected, for example, by screening or by symptoms. In such an analysis, it is important to note that survival times are influenced by several biases. One important factor is the bias due to the so-called lead time, i.e., the time by which the diagnosis has been brought forward by screening. Therefore, survival may appear to be longer even if earlier diagnosis did not affect disease progression. We examine whether there is a remaining survival difference after correction for bias that could be attributable to the mode of detection, for example, because of higher quality of care. These results can then be used to further improve the survival of breast cancer patients.