Education and lifestyle factors linked with complex mental activity are thought to affect the progression of cognitive decline. Collectively, these factors can be combined to create a cognitive reserve or cognitive lifestyle score. This study tested the association between cognitive lifestyle score and cognitive change in a population-based cohort of older persons from five sites across England and Wales. Data came from 13,004 participants of the Medical Research Council Cognitive Function and Ageing Study who were aged 65 years and over. Cognition was assessed at multiple waves over 16 years using the Mini-Mental State Examination. Subjects were grouped into four cognitive states (no impairment, slight impairment, moderate impairment, severe impairment) and cognitive lifestyle score was assessed as a composite measure of education, mid-life occupation, and current social engagement. A multi-state model was used to test the effect of cognitive lifestyle score on cognitive transitions. Hazard ratios for cognitive lifestyle score showed significant differences between those in the upper compared to the lower tertile with a more active cognitive lifestyle associating with: a decreased risk of moving from no to slight impairment (0.58, 95% CI (0.45, 0.74)); recovery from a slightly impaired state back to a non-impaired state (2.93 (1.35, 6.38)); but an increased mortality risk from a severely impaired state (1.28 (1.12, 1.45)). An active cognitive lifestyle is associated with a more favorable cognitive trajectory in older persons. Future studies would ideally incorporate neuroradiological and neuropathological data to determine if there is causal evidence for these associations.