Background & aims: Previous studies have reported that the Japanese diet is associated with a lower risk of dementia; however, whether changes in adherence to the Japanese diet affects incident dementia remains unknown. We aimed to evaluate the association between long-term changes in adherence to the Japanese diet and risk of incident dementia among older Japanese individuals. Methods: We collected dietary information from community-dwelling older individuals living in Ohsaki city, Japan using a validated 39-item food frequency questionnaire in 1994 and 2006. Adherence to the Japanese diet was assessed using the 8-item Japanese Diet Index (JDI8) score (range: 0 to 8 points). Changes in adherence to the Japanese diet were defined as changes in the JDI8 score from 1994 to 2006. Next, the participants were classified into five groups: great decrease, moderate decrease, no changes [ref.], moderate increase, or great increase. Then, 3146 Japanese adults aged >65 years in 2006 were followed-up for 5.7 years. Incident dementia was retrieved from the long-term care insurance database. The Cox proportional hazards model was used to estimate multivariable-adjusted hazard ratios (HRs) and 95% confidence intervals (95% CIs) for incident dementia. Results: During 14,336 person-years of follow up, 231 cases of dementia were ascertained. Compared with no changes in the JDI8 score, the multivariable-adjusted HRs (95%CIs) were 1.72 (1.13, 2.62) for great decrease, 1.10 (0.73, 1.66) for moderate decrease, 0.82 (0.54, 1.25) for moderate increase, and 0.62 (0.38, 1.02) for great increase (p-trend <0.0001). Conclusions: An increase in adherence to the Japanese diet was associated with a reduced risk of incident dementia, whereas a decrease in adherence was associated with an elevated risk among older Japanese individuals. (c) 2020 The Author(s). Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).