Cognitive changes during the menopausal transition: a longitudinal study in women with and without HIV.

OBJECTIVE To assess longitudinal changes in cognitive performance across menopause stages in a sample comprised primarily of low-income women of color, including women with HIV (WWH). METHODS A total of 443 women (291 WWH; 69% African American; 18% Hispanic; median age = 42 y) from the Women's Interagency HIV Study completed tests of verbal learning and memory, attention/working memory, processing speed, verbal fluency, motor skills, and executive function first at an index premenopausal visit and thereafter once every 2 years for up to six visits (mean follow-up = 5.7 y). General linear-mixed effects regression models were run to estimate associations between menopause stages and cognition, in the overall sample and in WWH. We examined both continuous scores and categorical scores of cognitive impairment (yes/no >1 standard deviation below the mean). RESULTS Adjusting for age and relevant covariates, the overall sample and WWH showed longitudinal declines in continuous measures of learning, memory, and attention/working memory domains from the premenopause to the early perimenopause and from the premenopause to the postmenopause, Ps < 0.05 to < 0.001. Effects on those same domains were also evident in categorical scores of cognitive impairment, with the increased odds of impairment ranging from 41% to 215%, Ps < 0.05 to < 0.001. The increase in predicted probability of impairment by menopausal stage (% affected) ranged from 4% to 13%. CONCLUSIONS Menopause stage was a key determinant of cognition in a sample of low-income women of color, including WWH. Many of these changes reached a clinically significant level of cognitive impairment.