Does befriending by trained lay workers improve psychological well-being and quality of life for carers of people with dementia, and at what cost? A randomised controlled trial.

Objectives: To determine whether a social support intervention (access to an employed befriending facilitator in addition to usual care) is effective compared with usual care alone. Also to document direct and indirect costs, and establish incremental cost-effectiveness.

Design: The Befriending and Costs of Caring (BECCA) trial was a cost-effectiveness randomised controlled trial. Data on well-being and resource use were collected through interviews with participants at baseline and at 6, 15 and 24 months.

Setting: This research was carried out in the English counties of Norfolk and Suffolk, and the London Borough of Havering. It was a community-based study.

Participants: Participants were family carers who were cohabiting with, or providing at least 20 hours' care per week for, a community-dwelling relative with a primary progressive dementia.

Interventions: The intervention was 'access to a befriender facilitator' (BF). BFs, based with charitable/voluntary-sector organisations, were responsible for local befriending schemes, including recruitment, screening, training and ongoing support of befriending volunteers, and for matching carers with befrienders. The role of befrienders was to provide emotional support for carers. The target duration for befriending relationships was 6 months or more.

Main Outcome Measures: Depression was measured by the Hospital Anxiety and Depression Scale (HADS) at 15 months postrandomisation. The health-related quality of life scale EQ-5D (EuroQol 5 Dimensions) was used to derive utilities for the calculation of quality-adjusted life-years (QALYs).

Results: A total of 236 carers were randomised into the trial (116 intervention; 120 control). At final follow-up, 190 carers (93 intervention; 97 control) were still involved in the trial (19% attrition). There was no evidence of effectiveness or cost-effectiveness from the primary analyses on the intention-to-treat population. The mean incremental cost per incremental QALY gained was in excess of 100,000 pounds, with only a 42.2% probability of being below 30,000 pounds per QALY gained. Where care-recipient QALYs were included, mean incremental cost per incremental QALY gained was 26,848 pounds, with a 51.4% probability of being below 30,000 pounds per QALY gained. Only 60 carers (52%) took up the offer of being matched with a trained lay befriender, and of these only 37 (32%) were befriended for 6 months or more. A subgroup analysis of controls versus those befriended for 6 months or more found a reduction in HADS-depression scores that approached statistical significance (95% CI -0.09 to 2.84).

Conclusions: 'Access to a befriender facilitator' is neither an effective nor a cost-effective intervention in the support of carers of people with dementia, although there is a suggestion of cost-effectiveness for the care dyad (carer and care recipient). In common with many services for carers of people with dementia, uptake of befriending services was not high. However, the small number of carers who engaged with befrienders for 6 months or more reported a reduction in scores on HADS depression that approached statistical significance compared with controls (95% CI -0.09 to 2.84). While providing only weak evidence of any beneficial effect, further research into befriending interventions for carers is warranted.