There are numerous ways to better integrate the elderly into communities, many of which are contingent upon whether they will remain in their pre-retirement homes or make a move. Using a life course perspective, this paper establishes that residential history, social and family relations, socio-economic status, and health trajectories measured at mid-life can be associated with moves in later life, either directly, or indirectly through their effect on the mid-life residential trajectory. These relationships are examined with multivariate Cox proportional hazards and Poisson regression models, using data from the Ontario Longitudinal Study of Aging. These findings suggest directions for future research, to aid the development of public policy for the large baby-boom cohorts that are just entering mid-life.
