Housing environments, social isolation, and mortality among low-income older renters in the United States.

Housing environments constitute a fundamental social determinant of health in later life, yet little is known about how different housing contexts shape survival among economically vulnerable older adults. This study examines whether mortality among low-income older renters follows a graded housing ecology and whether the consequences of social isolation are conditioned by residential context. Using twelve waves (2011-2022) of nationally representative data from the National Health and Aging Trends Study, we analyze 2790 low-income renters (9337 person-wave observations). Mortality risk was estimated using longitudinal discrete-time hazard models with weighting to address attrition and housing selection. A clear mortality gradient emerged across housing types. In fully adjusted models, residents of traditional rental housing exhibited significantly higher mortality hazards than those in subsidized senior housing (SSH) (HR = 1.56, p < .01), while non-subsidized congregate housing showed intermediate risk (HR = 1.32, p < .10). Social isolation independently predicted mortality (HR = 1.51 for isolated; HR = 1.41 for severely isolated, p < .05). However, the mortality consequences of isolation varied by housing context. Socially isolated residents in non-subsidized housing faced nearly double the mortality risk relative to socially integrated SSH residents (HR = 2.00, p < .05). Predicted probabilities were similar across housing types among socially integrated older adults (6-7%) but diverged under isolation, rising to 9-13% outside SSH. These findings suggest that housing environments function as meso-level social ecologies that structure exposure to relational risk. Structural supports embedded in subsidized senior housing appear to buffer the life-threatening consequences of isolation, whereas less institutionally embedded contexts amplify vulnerability in later life.