Socioeconomic correlates of social asymmetry in English older adults: Findings from a population-based cohort study.

OBJECTIVES: Discrepancies between loneliness and social isolation (i.e., social asymmetry) are linked to physical and psychological health in later life. While psychological correlates of social asymmetry are increasingly documented, socioeconomic correlates remain underexplored. The study examined whether individual- and area-level socioeconomic indicators were longitudinally associated with social asymmetry in English older adults. METHODS: Data were drawn from 8,164 older adults aged 60+ participating in Waves 7-9 (2014-2019) of the English Longitudinal Study on Ageing. Social asymmetry was operationalised as a continuous discrepancy score between standardised loneliness (UCLA-3 Loneliness) and social isolation (modified Berkman-Syme Social Network Index), with positive values indicating social resilience (i.e., lower loneliness relative to isolation) and negative values indicating social vulnerability (i.e., higher loneliness relative to isolation). Time-varying generalised estimating equations assessed longitudinal associations between social asymmetry and multiple socioeconomic indicators. RESULTS: All socioeconomic indicators were independently associated with social asymmetry, even after adjusting for psychological covariates. Third level education (β=.44, p<.001), higher household income (β=.22, p<.001) and subjective social status (β=.004, p<.001) were associated with social resilience. Conversely, individual-level deprivation (β=-.04, p<.001), area-level deprivation (β=-.20, p<.001), population density (β=-.18,p<.001), rare/no internet use (β=-.26, p<.001), unemployment (β=-.07, p<.05) and renting in either private (β=-.35, p<.001) or social housing (β=-.41, p<.001) were associated with social vulnerability. DISCUSSION: Social asymmetry is associated with both psychological factors and structural socioeconomic inequalities. Addressing loneliness in later life, when social networks can shrink, may therefore require interventions that extend beyond individual behavioural changes to target broader community-level socioeconomic inequalities.