Background: In animals, higher endogenous or exogenous corticosteroids cause neuronal dysfunction, damage, and loss, especially in the hippocampus. In humans, high cortisol levels have been linked to memory impairment.
Objective: To prospectively examine the relation between morning basal cortisol level and change in cognitive performance during an average follow-up of 2 years.
Design: Longitudinal cohort.
Setting: Geographic community in southern California.
Measurements: Between 1984 and 1987, blood for basal morning cortisol was obtained along with pertinent medical, behavioral, and physical covariates from 749 post-menopausal women, mean age 72 years, who were not taking corticosteroids or postmenopausal hormones. Cognitive function was assessed in 502 women in the period from 1988 to 1991 and repeated in 136 women in the period from 1991 to 1993.
Results: In both age- and multiply adjusted linear regression models, higher baseline cortisol was a significant predictor of worsened category fluency. No other cognitive domain tested was related to cortisol.
Conclusions: These results support the hypothesis that subtle dysregulation of cortisol axis is related to memory loss. Interventions that block this pathway may provide new therapeutic options to prevent cognitive decline.