Age and distance to death in the Seattle longitudinal study

A series of hierarchical regression models was used to determine if time to death was a significant independent variable for level and seven-year change in intellectual performance for 1,214 community-dwelling adults. Distance to death explained a significant amount of the variance of intellectual performance at individuals' last measurement but not of the decline in performance after controlling for age, education, gender, and survivorship. The inclusion of time to death improved the proportion of unique variance explained by about 1% to 3% and between 4% and 10.4% of the total variance explained. Decedents had lower levels of verbal meaning, spatial ability reasoning ability, and psychomotor speed at last measurements and greater amounts of seven-year decline on verbal meaning and psychomotor speed. The inclusion of distance to death may help improve the explanation of variability in performance associated with increased age.