The diagnosis of hallucination for unusual perceptions such as deathbed visions, near-death experiences, or visions of the bereaved, is unhelpful in palliative medicine both academically and clinically. This paper reviews the broad prevalence data about unusual perceptions in the general population as background to identifying the more narrow epidemiological source from which the much smaller focus on hallucinations seem to emerge. Major debates and limitations of current hallucination research are reviewed to show that current academic and clinical certainties are largely confined to unusual perceptions that can be readily linked to psychopathology, quite specific organic disease states and psychoactive drug use. Current state-of-the-art in hallucination studies does not warrant broad or uncritical use of this type of diagnosis in end-of-life care. Conclusions from interdisciplinary (as opposed to single discipline) hallucination studies suggest that the way forward for clinical and research work in palliative medicine may lie in a more biographical and cultural approach to unusual perceptions at the end of life.