Cognitive and sociodemographic risk factors for mortality in the seattle longitudinal study

Citation
Hb. Bosworth et al., Cognitive and sociodemographic risk factors for mortality in the seattle longitudinal study, J GERONT B, 54(5), 1999, pp. P273-P282
Citations number
64
Categorie Soggetti
Public Health & Health Care Science","Medical Research General Topics
Journal title
JOURNALS OF GERONTOLOGY SERIES B-PSYCHOLOGICAL SCIENCES AND SOCIAL SCIENCES
ISSN journal
10795014 → ACNP
Volume
54
Issue
5
Year of publication
1999
Pages
P273 - P282
Database
ISI
SICI code
1079-5014(199909)54:5<P273:CASRFF>2.0.ZU;2-I
Abstract
The relationship between cognitive function and survivorship was examined i n a community-dwelling sample. Survival analysis was used to examine how le vel and change in intellectual functioning, verbal memory, perceptual speed and psychomotor speed were related to mortality in a sample of 601 individ uals who subsequently died (decedents; n = 342 men; n = 259 women; M = 73.8 1 years of age) and a control group of 609 survivors (n = 296 men; n = 313 women; M = 71.96). The sample of survivors was selected to be of similar ag e and to have a similar level of education as the decedents. individuals in the lowest 25th percentile of performance (crystallized abilities, visuali zation abilities, verbal memory, and perceptual and psychomotor speed) had a significant risk for subsequent mortality compared to individuals in the highest 25th percentile. However, after adjusting for demographic variables and psychomotor speed, only perceptual speed remained a significant risk f actor for mortality Significant 7-year declines (lowest 25th percentile) in measurements of Verbal Meaning, Spatial Ability, Reasoning Ability, and Ps ychomotor Speed were risk factors for subsequent mortality relative to thos e who had the least amount of decline. The relationship between mortality a nd cognitive function tended to be a specific rather than a pervasive pheno menon, even after adjusting for sociodemographic factors and psychomotor sp eed. Decrease in cognitive performance tended to be a better predictor of s ubsequent mortality than was the level of cognitive performance.