Predictors of perceived social support and loneliness over time: A population-based study in Switzerland, 2021-2025

Objectives Identifying predictors of strong social connections is essential for the development of successful public health promotion strategies. However, longitudinal research on this topic is limited. As different measures of social connection may influence health via distinct pathways, we examined predictors of two measures – perceived social support and loneliness – over time. Study design Population-based longitudinal cohort study. Methods Data were from 6311 adults (mean age = 51 years [range 18–96 years], 58% women) assessed every year from 2021 to 2025. We used multivariate linear mixed-effects models to examine predictors of perceived social support and loneliness, with linear combinations of model coefficients (lincom) to test whether the effects differed between the two outcomes. Results Common predictors of (low) perceived support and (heightened) loneliness included living alone, poorer self-rated health, having fewer psychological resources, and adverse life events. There were differential effects of age, sex, education, and nationality, such that loneliness decreased with age, while male sex, lower education, and foreign nationality were associated with less perceived social support but not with loneliness. Over time, younger age and non-persistent exposure to adversity were associated with a small but significant improvement in perceived social support and loneliness, while stronger psychological resources were associated with an improvement in loneliness. Conclusions Our findings identify those who may particularly benefit from interventions aiming to enhance social connections. In particular, they suggest that mitigating adversity, and the development of strong psychological resources, can help to promote a sense of social connection.