Citation

Hall PA, Dubin JA, Crossley M, Holmqvist ME, D'Arcy C. Psychosom. Med. 2009; 71(2): 196-204.

Affiliation

Department of Kinesiology, University of Waterloo, 200 University Avenue West, Waterloo, ON N2L 3G1, Canada. pahall@healthy.uwaterloo.ca

Copyright

(Copyright © 2009, American Psychosomatic Society, Publisher Lippincott Williams and Wilkins)

DOI

10.1097/PSY.0b013e318190d7f0

PMID

19073749

Abstract

OBJECTIVE: To assess the robustness of the association between intelligence quotient (IQ) and mortality in older adults and to examine whether or not the association can be explained by more specific cognitive processes, including individual differences in executive functioning. METHODS: We examined the associations among Full Scale IQ, individual IQ subtest scores, and 10-year mortality among older community-dwelling, adult participants in the Canadian Study of Health and Aging, who were verified as disease and cognitive-impairment free at baseline via comprehensive medical and neurological evaluation (n = 516). Survival analysis including Cox proportional hazards regression models were used to examine mortality risk as a function of Full Scale IQ and its specific subcomponents. RESULTS: An inverse association was found between IQ and mortality, but this did not survive adjustment for demographics and education. The association between IQ and mortality seemed to be predominantly accounted for by performance on one specific IQ subtest that taps executive processes (i.e., Digit Symbol (DS)). Performance on this subtest uniquely and robustly predicted mortality in both unadjusted and adjusted models, such that a 1-standard deviation difference in performance was associated with a 28% change in risk of mortality over the 10-year follow-up interval in adjusted models. CONCLUSIONS: The association between IQ and mortality in older adults may be predominantly attributable to individual differences in DS performance.

Language: en