The number of people with dementia is increasing in Japan, and establishing evidence for preventing dementia is necessary. This study was a randomized controlled trial in cognitively normal community-dwelling older adults aged 65 to 85 with diabetes and/or hypertension. Participants were randomly assigned in a 1:1 ratio. The intervention group underwent 90min of group-based weekly physical exercise, cognitive training, nutritional counseling, and vascular risk management for 18 months. The primary endpoint was the change in a cognitive composite score calculated by averaging the z-scores of seven neuropsychological tests from baseline to 18 months. We randomly assigned 203 participants to two groups, and 178 (87.7%) completed the 18-month follow-up. There was a significant group difference in the cognitive composite score change at 18 months (mean difference 0.16, 95% confidence interval: 0.04 to 0.27; p=0.009). An 18-month multimodal intervention for older adults at risk of dementia could improve their cognitive function. The trial was registered in the Clinical Trial Registration System (UMIN000041938). Japan-Multimodal Intervention Trial for Prevention of Dementia (J-MINT) PRIME Tamba was a randomized controlled trial to prevent dementia. We provided a multifactorial intervention based on the Finnish Geriatric Intervention Study to Prevent Cognitive Impairment and Disability (FINGER) trial methodology. The primary outcome, the cognitive composite score, improved with our intervention. Executive function/processing speed and memory improved in the intervention group. Intervention adherence was high, and no serious adverse events occurred.