We used a retrospective cohort study design to test the hypothesis of a relation between old-age depression and subsequent dementia. The study sample comprised 19103 patients aged 50 or more and born after 1910, included in a family-practice-based registration network. We estimated odds ratio (OR) and 95% confidence interval (95% CI) for a diagnosis of dementia in patients with or without previous late-onset depression and survival analysis, including hazard ratios resulting from Cox regression analysis. The OR for a diagnosis of dementia subsequent or not to late-onset depression was found at survival analysis: p = 0.26 (log rank test). Hazard ratio for patients with and without previous old age depression and subsequent dementia in patients aged 50 or more and born after 1910. This supports the hypothesis of old-age depression being a predictor, and possibly a causal factor, of subsequent dementia.