How effective are volunteers at supporting people in their last year of life? A pragmatic randomised wait-list trial in palliative care (ELSA)

Catherine Walshe, Steven Dodd, Matt Hill, Nick Ockenden, Sheila Payne, Nancy Preston, Guillermo Perez Algorta

Clinical care alone at the end of life is unlikely to meet all needs. Volunteers are a key resource, acceptable to patients, but there is no evidence on care outcomes. This study aimed to determine whether support from a social action volunteer service is better than usual care at improving quality of life for adults in the last year of life. A pragmatic, multi-centre wait-list controlled trial, with participants randomly allocated to receive the volunteer support intervention either immediately or after a 4 week wait. Trained volunteers provided tailored face-to-face support including befriending, practical support and signposting to services, primarily provided within the home, typically for 2-3 hours per week. The primary outcome was rate of change of quality of life at 4 weeks (WHO QOL BREF, a general, culturally sensitive measure). Secondary outcomes included rate of change of quality of life at 8 weeks and Loneliness (De Jong Gierveld Loneliness Scale), social support (mMOS-SS), and reported use of health and social care services at 4 and 8 weeks. In total, 196 adults (61% (n = 109) female; mean age 72 years) were included in the study. No significant difference was found in main or secondary outcomes at 4 weeks. Rate of change of quality of life showed trends in favour of the intervention (physical quality of life domain: b = 3.98, CI, -0.38 to 8.34; psychological domain: b = 2.59, CI, -2.24 to 7.43; environmental domain: b = 3, CI, -4.13 to 4.91). Adjusted analyses to control for hours of volunteer input found significantly less decrease in physical quality of life in the intervention group (slope (b) 4.43, CI, 0.10 to 8.76). While the intervention also favoured the rate of change of emotional (b = -0.08; CI, -0.52 to 0.35) and social loneliness (b = -0.20; CI, -0.58 to 0.18), social support (b = 0.13; CI, -0.13 to 0.39), and reported use of health and social care professionals (b = 0.16; CI, -0.22 to 0.55), these were not statistically significant. No adverse events were reported. Clinicians can confidently refer to volunteer services at the end of life. Future research should focus on 'dose' to maximise likely impact. The trial was prospectively registered. ISRCTN Registry: ISRCTN12929812 , registered 20 May 2015.