Laura M Holdsworth, Heather Gage, Simon Coulton, Annette King, Claire Butler

Rapid response services operating 24 h a day have been advocated in UK health policy to support dying patients at home, though there is limited evidence of their effectiveness. To assess the impact of a rapid response hospice at home service (intervention) on people dying in their preferred place, and carer quality of life, compared to usual care (control). Quasi-experimental multi-centred controlled evaluation. Patient data were collected from hospice records; carers completed postal questionnaires to report quality of life, anxiety and depression. Community served by one hospice (three contiguous sites) in South East England; 953 patients who died with a preferred place of death recorded and 64 carers who completed questionnaires. There was no significant difference between control and intervention groups in proportions achieving preferred place of death (61.9% vs 63.0% (odds ratio: 0.949; 95% confidence interval: 0.788-1.142)). People living at home alone were less likely to die where they wanted (0.541; 95% confidence interval: 0.438-0.667). Carers in the intervention group reported worse mental health component summary scores (short form-12, p = 0.03) than those in the control group; there were no differences in other carer outcomes. The addition of a rapid response hospice at home service did not have a significant impact on helping patients to die where they wanted in an area already well served by community palliative care. Recording preferences, and changes over time, is difficult and presented challenges for this study.