Objective Most terminally ill patients and relatives prefer care and death to occur at home. However, in many cases patients die in hospital and the question arises whether social inequity in palliative care exists. The aim of this study was to analyse associations between dying at home and demographic and socioeconomic characteristics adjusted for healthcare utilisation among Danish cancer patients.

Method Population-based, cross-sectional register study in Aarhus County, Denmark. 599 deceased adults who died from cancer from 1 March to 30 November 2006 in a well-defined geographical area were identified. Based on unique personal identifier numbers, socioeconomic data and healthcare utilisation from different registers were retrieved and analysed.

Results Multivariate analysis showed that dying at home was negatively associated with, first, being either 50–59 (prevalence ratio (PR): 0.67 (95% CI 0.45 to 0.99)) or 70–79 years of age (PR: 0.83 (95% CI 0.70 to 0.99)) compared with being 80 years or above; second, a middle personal income compared with a high income (PR: 0.86 (95% CI 0.75 to 1.00)); and, third, being employed or having a leadership position compared with being unemployed/student/receiving social security (PR: 0.72 (95% CI 0.53 to 0.98)).

Conclusion The found socioeconomic differences in whether death occurred at home or at institutions indicate that age, income and social class must be taken into account when palliative care services engage in fulfilling preferences of dying at home. This may lead to more equality in the possibility of dying at home, despite differences in socioeconomic level.