Bone mineral density and verbal memory impairment: Third National Health and Nutrition Examination Survey.

Previous studies have examined the relation of endogenous estrogen levels or estrogen replacement therapy to the risk of poor cognitive function, but results have been inconclusive. Bone mineral density has been proposed as a marker for cumulative estrogen exposure. The authors studied the relation of bone mineral density to the prevalence of verbal memory impairment among 4,304 elderly subjects in the Third National Health and Nutrition Examination Survey (1988-1994). Bone mineral density was measured in five regions of the proximal femur with dual-energy x-ray absorptiometry. Verbal memory was assessed using delayed recall of a three-item word list and a six-item story. Verbal memory impairment was defined as a combined score of <4. The prevalence of verbal memory impairment for each increasing bone mineral density quintile at the femoral neck was 8.35, 5.74, 5.22, 5.00, and 3.38% in women and 11.54, 7.27, 8.47, 6.29, and 5.89% in men, respectively. With adjustment for age, sex, and other covariates, the prevalence ratios of verbal memory impairment for each increased bone mineral density quintile were 1.00, 0.64, 0.65, 0.55, and 0.44, respectively (p for trend < 0.001). These results suggest that bone mineral density in the elderly is associated with verbal memory impairment. The mechanisms underlying this relation are not understood, but cumulative exposure to estrogen may play a role.