The complexity and nuance of how social networks shape dietary behaviours and health dynamics remain underexplored, particularly in collectivist societies where family and peer relationships strongly impact health. This study applies social network analysis to examine these dynamics in Singapore.

An online household survey of young adults (age 21–35) and family (21+) assessed the consumption of healthy food groups (fruit, vegetable intake), unhealthy food groups (fast food, snack consumption) and social network characteristics (interaction frequency, emotional closeness, shared meals and perceived health influence). Data were analysed using network analysis, mixed regression models and generalised estimating equations.

Online Singaporean household survey.

Among 116 participants from thirty-six households, 345 unique individuals and 1145 dyadic relationships were identified, with networks averaging 9·7 nodes (sd: 4·7) and 33·2 edges (sd: 27·3). Mutual health influence was strongest in spousal (β = 0·89, 95 % CI: 0·42, 1·35) and intergenerational ties (older-to-younger: β = 0·62, 95 % CI: 0·29, 0·94; younger-to-older: β = 0·36, 95 % CI: 0·03, 0·68) and associated with emotional closeness (β = 0·38, 95 % CI: 0·30, 0·46) and shared meals (β = 0·43, 95 % CI: 0·36, 0·49). Greater family health effort correlated with lower snack (Adjusted Odds Ratio [AOR]: 0·50, 95 % CI: 0·29, 0·85) and fast-food consumption (AOR: 0·41, 95 % CI: 0·22, 0·77), while higher perceived family health associated with increased snack intake (AOR: 3·21, 95 % CI: 1·58, 6·52). Frequent meals with friends associated with lower fast-food intake (AOR: 0·50, 95 % CI: 0·30, 0·84), but no associations with fruit or vegetable intake were found.

Findings highlight intergenerational and spousal ties as key health influencers, particularly through shared meals, and the complex role of social networks in shaping diet. Analyses suggest network-based interventions may be more useful in reducing unhealthy rather than promoting healthy eating behaviours.