A meta-analysis of behaviour change techniques in social interventions targeting improved cognitive function in older adults

Background: Limited social networks in older adults are linked with increased dementia risk. However, there is a lack of knowledge on whether socially-based behavioural interventions (i.e., programs designed to increase individual’s social opportunities, engagement or networks) can improve cognitive function, as well as the role of applied behaviour change techniques (BCTs) in effective interventions. This systematic review and meta-analysis aimed to (i) quantify the effectiveness of social-based behavioural interventions in improving cognition in older adults, and (ii) identify which BCTs increase social activity behaviour of older adults. Methods: Six electronic databases were searched with restrictions for age (>65 years) and English language from inception to July 2023 (PROSPERO:CRD42021283382) for articles reporting social-based behavioural randomised controlled trials and using a measured outcome of cognitive function. Behaviour change techniques were mapped to the BCT V1 model and risk of bias was assessed. Pooled effect sizes from eligible studies were synthesised using RevMan. Results: We identified 9528 records and included 15 studies (N=1785 participants). Meta-analyses showed that social-based interventions had a medium effect on global cognition (d=0.80, 95% CI 0.58 to 1.02, p<0.01), but not executive function. The most frequently used intervention components were social-based communication (e.g., chatting, boosting social engagement), group arts-based tasks (e.g., knitting, music, craft) and guided reminiscence. The BCT demonstration of behaviour predicted significant cognitive effects and explained 94.6% of inter-study variation. Discussion: Findings carry implications for developing comprehensive strategies to promote social initiatives supporting cognitive health, particularly in addressing the challenges faced by older adults.