Abstract
BACKGROUND: Loneliness is a public health epidemic in the United States (US), with older adults being vulnerable to experiencing loneliness. Predictors of loneliness are less understood among racial/ethnic groups of US older adults, and few studies have included perceived institutional discrimination (PID), stressful life events (SLE), and perceived neighborhood characteristics (PNC) as antecedent stressors of loneliness in diverse older adult samples. Our study assessed the relationship between these stressors and loneliness among specific racial/ethnic groups of older adults. METHODS: We used the Health and Retirement Study data (n = 9,904) to examine whether PID, SLE, and PNC were associated with loneliness. Loneliness was measured using the 11-item UCLA Loneliness Scale. PID included unfairly not hired for a job, unfairly prevented from moving into a neighborhood, and unfairly treated by the police. SLE included moving to a worse neighborhood/residence, being robbed or burglarized, and unemployed/looking for a job. PNC were measured as discohesion and disorder. Lagged multivariate linear regression models regressed loneliness (2014/2016 HRS waves) on PID, SLE and PNC (2010/2012 HRS waves) measured as cumulative totals and individual items. Models were stratified by Black (BOAs), Hispanic/Latinx (HOAs), and White (WOAs) older adults. RESULTS: Cumulative totals of PID, SLE, and neighborhood discohesion were associated with loneliness among BOAs while only discohesion was associated with loneliness among HOAs. Cumulative totals for PID, SLE, and PNC were associated with loneliness among WOAs. Individual stressors predicting loneliness for BOAs were moving to a worse residence and being robbed/burglarized. For HOAs, being prevented from moving to a neighborhood was associated with greater loneliness while being robbed/burglarized was associated with less loneliness. Individual stressors predicting greater loneliness for WOAs were being unfairly not hired for a job, receiving unfair treatment during police encounters, and moving to a worse residence. CONCLUSIONS: Our study finds racial/ethnic variation in psychosocial stressors predicting loneliness four years later. Nevertheless, neighborhood discohesion was the most salient stressor and was associated with greater loneliness across all racial/ethnic groups. Future research and interventions should consider the differing stress appraisal processes across groups and to support the development of resources and policies to ameliorate loneliness among diverse older adults.