Objective This study investigates the independent and combined potential of slowed gait speed and slowed processing speed as predictors of adverse health outcomes. The role of depressive symptoms in these associations is also investigated. Methods In the prospective cohort study, using the Longitudinal Aging Study Amsterdam database, three study samples for each outcome variable were defined: persistent cognitive decline (PCD; N = 1,271, 13 years of follow-up), falls (N = 1,282, 6 years of follow-up), and mortality (N = 1,559, age 74.9 ± 5.8, 21 years of follow-up). At baseline, gait speed (6-m walk with a turn at 3 m), processing speed (coding task), depressive symptoms (Center for Epidemiologic Studies Depression Scale), and basic demographic data were assessed. Also, time to PCD, falls, and mortality were assessed. Cox (for PCD and mortality) and stratified Cox (for falls) regression models were used. Results Slowed processing speed predicted PCD (HR: 7.8; 95% CI: 3.3–18.8), slowed gait speed predicted falls (HR: 1.3; 95% CI: 1.0–1.5), and both measures predicted mortality (gait speed HR: 2.1; 95% CI: 1.6–2.6; processing speed HR: 1.9; 95% CI: 1.6–2.4). Each association remained significant after adjusting for the other slowing symptom. Slowed processing speed only predicted falls in the presence of slowed gait (interaction). A slowing sum score that combines both slowing symptoms predicted all three outcomes. The associations were not influenced by depressive symptoms. Conclusion Slowing of thought is as relevant as slowing of movement to predict adverse health outcomes, because they seem to represent separate underlying pathologies.