Cognitive Performance and Age: Norms from the Maine-Syracuse Study

Abstract

The primary objective of this study was to provide contemporary normative data on aging and cognition from an ongoing community-based study. This dementia and stroke-free sample (age range = 20–79; mean = 53) consisted of 623 women and 322 men participating in the Maine-Syracuse Longitudinal Study at waves 4 to 6 (1993 to 2003). We employed a battery of 22 widely utilized cognitive tests. A 5 (age) × 3 (education) × 2 (gender) analysis of variance indicated that, in general, higher educated and younger participants exhibited better performance on cognitive tests. We found education group to be the strongest, and gender to be the weakest, predictor of cognitive performance. However, education cohort was not significantly associated with every cognitive outcome, nor was age cohort membership. The addition of cardiovascular disease health variables to a model including age, education, and gender groupings provided statistically significant, but modest, increases in prediction of performance on some tests. Results are discussed in relation to findings for previous studies presenting normative data on cognitive ability as a function of age, education, and gender.

This work was supported by National Institute on Aging project AG03055 and National Heart, Lung, and Blood Institute projects HL67358 and HL81290. The authors wish to thank Amanda Goodell and Anna Sweeney for their assistance in the editing of the manuscript.

Notes

Note. Demographic variables, cardiovascular risk factors and diseases are defined in Methods.

¹Minorities include: African Americans, American Indians, Alaskan Natives, non-white Hispanics, and Asian/Pacific Islanders.

²Depressed Mood based on a Zung Self-Rating Depression Scale of 50 or more (Carroll et al., 1973).

³Hypertension is defined as SBP of 140 mm Hg or more, DBP of 90 mm Hg or more, or treatment with antihypertensive medications.

⁴Obesity is defined as a body mass index of 30 or more.

†Possible value not determinable by the test design.

¹Derived score.

²Trail Making Difference = Trail Making B–Trail Making A.

³Wechsler Adult Intelligence Scale (Wechsler, 1955).

⁴Halstead-Reitan Battery (Reitan & Wolfson, 1993).

⁵Wechsler Memory Scale—Revised (Wechsler, 1987).

⁶Auditory Verbal Learning Test (Rey, 1964).

⁷Multilingual Aphasia Exam (Benton & Hamsher, 1976).

⁸Lezak (1995 2004).

⁹MMSE (Folstein, Folstein, & McHugh 1975).

Note. Age, education, and gender were controlled for each other and the two- and three-way interactions.

∗p < .05; ∗∗p < .01; ∗∗p < .001.

¹AVLT = Rey Auditory Verbal Learning Test.

¹HV = health variables (all yes/no; MI, diabetes mellitus, TIA, CHD, hypertension, smoker, obesity).

²Zung = depressive symptoms (score on the Zung Self-Rating Depression Scale).

³AVLT = Rey Auditory Verbal Learning Test.

∗p < .05

∗∗p < .01

∗∗∗p < .001.