Gran, Jon MichaelStensrud, Mats Julius2024-05-162024-05-162024-05-162024-01-23https://infoscience.epfl.ch/handle/20.500.14299/207924WOS:001199039700021In survival studies, all participants are rarely followed until they die. Methods for lifetime analysis are designed to handle such incompletely observed lifetimes. We call these lifetimes censored. The standard method for lifetime analysis assumes that the censoring times are independent of the lifetimes. When is this assumption broken? And what do we do then?Life Sciences & Biomedicinetext::journal::journal article::research article