P1913 Does neighbourhood cohesion protect against chronic loneliness for all older adults?: A longitudinal analysis of data from UK Household Longitudinal Study

Abstract

Background: Risk factors for loneliness operate at micro, meso and macro level. Evidence about loneliness risk factors is dominated by individual level data. Evidence about meso level factors is sparse and research rarely differentiates different loneliness trajectories. We address these gaps by investigating the role of neighbourhood cohesion as a ‘protective’ factor against chronic loneliness and focus upon three groups of older adults under-represented in research: minority ethnic people, those who identify as lesbian, gay and bisexual (LGB) and those living with a longstanding disability. Study Setup (Methods): We used data from 10,916 community-dwelling adults aged 50 and over who participated in four waves of the UK Household Longitudinal Study. Loneliness was measured using the three-item UCLA scale and neighbourhood cohesion by the Buckner scale. We define four loneliness trajectories: not lonely, transient (lonely once), fluctuating (two or three occasions of loneliness) and chronic (lonely at every wave). Results: Neighbourhood cohesion was highest for respondents who identified as Bangladeshi, Pakistani and white Irish and lowest for participants with mixed ethnic backgrounds. Multinomial regression models indicate that following adjustment for key socio-demographic factors and mental wellbeing, better neighbourhood cohesion was associated with less transient, fluctuating and chronic loneliness for the overall sample. For those ageing with disability neighbourhood cohesion was protective for fluctuating and chronic loneliness but no evidence of this was observed for ethnic minority populations or those who identified as LGB. Discussion: These findings highlight how neighbourhood level factors are experienced differentially across different groups of older adults.