The influence of general practitioner community hospitals on the place of death of cancer patients.

All deaths from cancer were identified from death certificates in the Exeter Health District for a period of one year. Place of death, age, cancer type and access to general practitioner community hospital beds and the domiciliary hospice service were recorded. There were 1022 deaths attributable to cancer (parts 1a, 1b or 1c of the death certificate) who were patients of general practitioners in the health district. The place of death for patients with access to community hospital beds were: home 173/590 (29%), community hospital 232/590 (39%), specialist services unit 102/590 (17%), nursing or residential home 32/590 (5%), Marie Curie hospice 51/590 (9%). For patients without access to community hospital beds the place of death was: home 177/427 (41%), specialist service unit 165/427 (39%), nursing or residential home 42/427 (10%), Marie Curie hospice 43/427 (10%). The presence of community hospital beds was associated with a significant reduction of deaths in the specialist service unit (p < 0.001) and with a smaller reduction in home deaths (p < 0.01). Access to the domiciliary hospice services in areas with community beds was not associated with any significant change in the place of death. General practitioners cared for 74% of cases at the time of death in areas with access to community hospital beds and for 51% of cases without such access, which was a significant difference (p < 0.001). It therefore appears that community hospitals play a major role in the terminal care of cancer patients and access to such beds is associated with a decrease in cancer deaths occurring in specialist services beds.