Abstract
BACKGROUND: living alone in later life has been linked to psychological distress but less is known about the role of the transition into living alone and the role of social and material resources. METHODS: a total of 21,535 person-years of data from 4,587 participants of the British Household Panel Survey aged 65+ are analysed. Participants provide a maximum 6 years' data (t0-t5), with trajectories of living arrangements classified as: consistently partnered/ with children/alone; transition from partnered to alone/with children to alone. General Health Questionnaire (GHQ)-12 caseness (score >3) is investigated using multi-level logistic regression, controlling for sex, age, activities of daily living, social and material resources. RESULTS: after a transition from partnered at t0 to alone at t1, the odds for GHQ-12 caseness increased substantially, but by t3 returned to baseline levels. The odds for caseness at t0 were highest for those changing from living with a child at t0 to living alone at t1 but declined following the transition to living alone. None of the covariates explained these associations. Living consistently alone did confer increased odds for caseness. CONCLUSIONS: living alone in later life is not in itself a strong risk factor for psychological distress. The effects of transitions to living alone are dependent on the preceding living arrangement and are independent of social and material resources. This advocates a longitudinal approach, allowing identification of respondents' location along trajectories of living arrangements.