© 2018 The Author(s). Background: Conflicting data in the literature raise the question whether gout, independent of its treatment, increases the risk of dementia in the elderly. Our objective was to assess whether gout in older adults is associated with the risk of incident dementia. Methods: We used the 5% Medicare claims data for this observational cohort study. We used multivariable-adjusted Cox proportional hazard models to assess the association of gout with a new diagnosis of dementia (incident dementia), adjusting for potential confounders/covariates including demographics (age, race, sex), comorbidities (Charlson-Romano comorbidity index), and medications commonly used for cardiac diseases (statins, beta-blockers, diuretics, and angiotensin converting enzyme (ACE)-inhibitors) and gout (allopurinol and febuxostat). Results: In our cohort of 1.71 million Medicare beneficiaries, 111,656 had incident dementia. The crude incidence rates of dementia in people without and with gout were 10.9 and 17.9 per 1000 person-years, respectively. In multivariable-adjusted analyses, gout was independently associated with a significantly higher hazard ratio of incident dementia, with a HR of 1.15 (95% CI, 1.12, 1.18); sensitivity analyses confirmed the main findings. Compared to age 65 to < 75 years, age 75 to < 85 and ≥ 85 years were associated with 3.5 and 7.8-fold higher hazards of dementia; hazards were also higher for females, black race or people with higher medical comorbidity. Conclusion: Gout was independently associated with a 15% higher risk of incident dementia in the elderly. Future studies need to understand the pathogenic pathways involved in this increased risk.