Cardiometabolic measures and cognition in early menopause - Analysis of baseline data from a randomized controlled trial

Lubna Pal, Kelly Morgan, Nanette F. Santoro, Jo Ann E. Manson, Hugh S. Taylor, Virginia M. Miller, Eliot A. Brinton, Rogerio Lobo, Genevieve Neal-Perry, Marcelle I. Cedars, S. Mitchell Harman, Taryn T. James, Carey E. Gleason

Research output: Contribution to journalArticlepeer-review


Objective: The relationships between cardiometabolic indices and cognition were examined in recently menopausal women. Methods: Cross-sectional analysis of baseline data from the KEEPS (Kronos Early Estrogen Prevention Study)-Cognitive ancillary study (n = 621). Cognitive performance was assessed by the Modified Mini Mental Status (3MS) score (primary outcome). Physical cardiometabolic indices included body mass index (BMI), waist circumference (WC), waist-to-hip ratio (WHR), and blood pressure (BP). Biochemical cardiometabolic indices included serum levels of high sensitivity C-reactive protein (hs-CRP), total cholesterol (TC), low-density lipoprotein (LDL-C), high-density lipoprotein (HDL-C), non-HDL (non-HDL-C), triglycerides (TG), fasting serum glucose (FSG), and insulin resistance (HOMA-IR). Socio-demographic variables included age, race/ethnicity, education, and lifestyle (physical activity, smoking). Central adiposity was defined as WC > 88 cm (>35 in) and WHR > 0.8. Separate stepwise multivariable analyses (GLM, ordinal logistic regression and logistic regression) assessed relationships between 3MS scores (as continuous, in tertiles and dichotomized at 90 respectively) with the measures of central adiposity (predictor variables); socio-demographic variables (age, time since menopause, race, educational status and lifestyle) and cardiometabolic variables (BP, lipids, FSG, HOMA-IR and hs-CRP) were examined as covariates. The final multivariable models included time since menopause, race, ethnicity, educational status, strenuous exercise, BMI ≥30 kg/m2, non-HDL-C and hs-CRP as covariates. Due to the high collinearity between the two indices of central adiposity, within each analytic strategy, separate models examined the respective associations of WC > 88 cm and WHR > 0.8 with 3MS score. Results: On adjusted analyses, indices of central adiposity were independent predictors of significantly lower 3MS scores (p < 0.05). Consistency in this relationship was observed across the three different multivariable regression analytic approaches (GLM, ordinal and logistic regression). Conclusions: Among recently menopausal women, WC > 88 cm and WHR > 0.8 were associated with significantly lower cognitive function, as reflected by lower 3MS scores. The mechanisms that might explain the observed negative implications of central adiposity for cognitive function warrant further study.

Original languageEnglish (US)
Pages (from-to)58-65
Number of pages8
StatePublished - Aug 2022


  • Central adiposity
  • Cognition
  • Menopause
  • Obese
  • Waist circumference
  • Waist:hip ratio

ASJC Scopus subject areas

  • Biochemistry, Genetics and Molecular Biology(all)
  • Obstetrics and Gynecology


Dive into the research topics of 'Cardiometabolic measures and cognition in early menopause - Analysis of baseline data from a randomized controlled trial'. Together they form a unique fingerprint.

Cite this