Objective: To examine the association between diabetes in midlife (1963- 196 8) and dementia more than three decades later (1999- 2001). Methods: The author s characterized dementia using standard methods for 1,892 participants among 2,6 06 survivors of 10,059 participants in the Israeli Ischemic Heart Disease study, a longitudinal investigation of the incidence of and risk factors for cardiovas cular disease among Jewish male civil servants in Israel. Face to face interview s were conducted with the 652 subjects identified as possibly demented by the Mo dified Telephone Interview for Cognitive Status. Logistic regression analysis wa s performed to assess the association of diabetes with dementia controlling for sociodemographic and cardiovascular variables compared to those with no cognitiv e impairment. Results: Of 1,892 assessed subjects (mean age 82 at assessment), 3 09 (16.3% ) had dementia. Diabetic subjects had significantly more dementia tha n non-diabetic subjects (χ 2 = 7.54, df = 1, p = 0.006, OR 2.83 [95% CI = 1 .40 to 5.71]). Those who survived to the time of this study were younger and hea lthier than those who died. Conclusions: Evidence for diabetes as a risk factor for dementia was found, similar to other epidemiologic studies. In contrast to t he earlier studies, however, the authors linked diabetes in midlife to dementia more than three decades later in the very old survivors of a large male cohort. Objective: To examine the association between diabetes in midlife (1963- 196 8) and dementia more than three decades later (1999- 2001). Methods: The author’s characterized use of standard methods for 1,892 participants among 2,606 survivors of 10,059 participants in the Israeli Ischemic Heart Disease study, a longitudinal investigation of the incidence of and risk factors for cardiovas cular disease among Jewish male civil servants in Israel. Face to face interview were were with the 652 subjects identified as possibly demented by the Mo dified Telephone Interview for Cognitive Status. Logistic regression analysis wa s performed to assess the association of diabetes with dementia control for sociodemographic and cardiovascular variables compared to those with no cognitiv e impairment. Results: Of 1,892 assessed subjects, mean age 82 at assessment, 3 09 (16.3%) had dementia. Diabetic subjects had significantly more dementia tha n non-diabetic subjects (χ 2 = 7.54 , df = 1, p = 0.006, OR 2.83 [95% CI = 1.40 to 5.71]). Those who survived to the time of this study were younger and hea lthier than those who died. Conclusions: Evidence for diabetes as a risk factor for dementia was found, similar to other past miracle studies. However, the authors linked diabetes in midlife to dementia more than three decades later in the very old survivors of a large male cohort.