Does Baseline Depression Predict Dementia Onset among the Old?: A 4 year Follow-up Study from the AGES Project

Background: Both depression and dementia are common among the old. However, studies which clarified the association between baseline depression and dementia are relatively few. Methods: To investigate such association, we used the AGES (Aichi Gerontological Evaluation Study) data. Subjects were functionally independent elderly (65+) at baseline (N=13,295) in 6 communities. They were followed from 2003 to 2007 for dementia onset. Dementia onset was determined according to the criterion (based on face to face interview and physician diagnosis) used in Japan's Public Long-Term Care Insurance System. Depression was assessed by GDS-15. Cox hazard proportional model stratified by sex was employed to calculate hazard ratios for dementia onset. Results: Out of 13,295 subjects, 225 men (3.5%) and 324 women (4.8%) developed dementia during the follow-up. Dementia onset increased exponentially with age. In Cox models, even after adjustment for age and illnesses, depression at baseline significantly predicted dementia onset among men with hazard ratios of 1.70 (p< .01) for mild depression and 2.95 (p< .001) for severe depression. Among women, such hazards were 1.89 (p< .001) and 2.40 (p< .001), respectively. To test is such association is due to diminished intellectual or social activities, we added instrumental activities of daily living (IADL) scores in the model and found depression-dementia association diminished. Conclusions: Baseline depression predicted dementia onset among the old in Japan. However, IADL scores were even stronger predictor of dementia, meaning that maintaining IADL might play important roles in delaying or preventing dementia.