Transitions in cognitive status among the aged in Japan.

This study examines the competing risk of cognitive impairment, mortality and study attrition over a three year period within a national probability sample of Japanese elderly (n = 1506). Younger age and fewer chronic conditions were related to recovery, while older age, being married, poorer self-rated health and depression were related to mortality. Impaired, urban respondents were more likely to drop out of the study than impaired rural respondents. For those 'intact' at baseline, the probabilities of impairment, death and non-response were 7, 6 and 16%. Older, less educated individuals were more likely to become impaired; older males, less educated, married, those in poorer self-rated health with poor functional health were more likely to die; and younger, single, urban living individuals with poor self-rated and functional health, a past smoking history and high levels of depression were the most likely to drop out of the study. A Japanese elder aged 65 is expected to spend about 14.6 years (81%) free from cognitive impairement and about 3.45 years (19%) experiencing some degree of cognitive impairment throughout the remaining lifetime.