Objectives To determine the relationships between walking speed, cognitive function, and the interaction between changes in these measures and dementia risk. Design Longitudinal observational study. Setting English Longitudinal Study of Ageing. Participants Individuals aged 60 and older (N=3,932). Measurements Walking speed and cognition were assessed at Waves 1 (2002–03) and 2 (2004–05) of the English Longitudinal Study of Ageing. New dementia cases were assessed from Wave 3 (2006–07) to Wave 7 (2014–15). The associations were modelled using Cox proportional hazards regression. Results Participants with faster baseline walking speeds were at lower risk of developing dementia (hazard ratio (HR)=0.36, 95% confidence interval (CI)=0.22–0.60). Those with a greater decline in walking speed from Wave 1 to 2 were at greater risk of developing dementia (HR=1.23, 95% CI=1.03–1.47). Participants with better baseline cognition (HR=0.42, 95% CI=0.34–0.54) were at lower risk of developing dementia. Those with a greater decline in cognition from Wave 1 to 2 were at greater risk of developing dementia (HR=1.78, 95% CI=1.53–2.06). Change in walking speed and change in cognition did not have an interactive effect on dementia risk (HR=1.01, 95% CI=0.88–1.17). Conclusion In this community‐dwelling sample of English adults, those with slower walking speeds and a greater decline in speed over time were at greater risk of developing dementia independent of changes in cognition. Further research is required to understand the mechanisms that may drive these associations.