Surveys in different countries (e.g. the UK, Belgium and The Netherlands) show a marked recent increase in the incidence of continuous deep sedation at the end of life (CDS). Several hypotheses can be formulated to explain the increasing performance of this practice. In this paper we focus on what we call the ‘natural death’ hypothesis, i.e. the hypothesis that acceptance of CDS has spread rapidly because death after CDS can be perceived as a ‘natural’ death by medical practitioners, patients' relatives and patients. We attempt to show that the label ‘natural’ cannot be unproblematically applied to the nature of this end-of-life practice. We argue that the labeling of death following CDS as ‘natural’ death is related to a complex set of mechanisms which facilitate the use of this practice. However, our criticism does not preclude the view that CDS may be clinically and ethically justified in many cases.