BACKGROUND: Befriending is a popular way in which to intervene to combat loneliness and social isolation among older people. However, there is a need to improve our understanding about how these interventions work, for whom and in which contexts, to make the best use of the increasing investment in the provision and delivery of befriending services.
METHODS: A realist evaluation was undertaken as it focuses on uncovering causal processes and interactions between mechanisms and contextual characteristics. Five case studies of befriending programmes in Northern Ireland were studied, reflecting variation in contextual variables, service user and provider characteristics. Data was collected via service documentation and semi-structured interviews (n = 46) with stakeholders involved in the delivery and receipt of befriending interventions.
RESULTS: Eight initial programme theories were generated, which were 'tested' in the case study analysis to uncover context-mechanism-outcome relationships. Mechanisms identified included reciprocity, empathy, autonomy, and privacy which were triggered in different contexts to support the alleviation of loneliness and social isolation. Reciprocity was 'triggered' in contexts where service users and befrienders shared characteristics, the befriender was a volunteer and befriending took the form of physical companionship. Contexts characterised in terms of shared experiences between befriender and service user triggered empathy. Autonomy was triggered in contexts where befriending relationships were delivered long-term and did not focus on a pre-defined set of priorities. Privacy was triggered in contexts where service users had a cognitive/sensory impairment and received one-to-one delivery.
CONCLUSION: This study improves understanding about how and why befriending interventions work. Findings indicate that services should be tailored to the needs of service users and take into consideration characteristics including mobility, impairments e.g. physical, sensory and/or cognitive, as well as the influence of service characteristics including payment for befrienders, fixed/long-term befriending relationship, one-to-one support and the impact of non-verbal communication via face-to-face delivery.