Abstract

This study examines longitudinal changes in sleep health across stages of the menopausal transition using repeated observations from a large cohort of midlife women. Different sleep health variables were measured as an ordinal outcomes. Because observations were recorded across multiple visits for the same individuals, the data exhibit within-subject correlation and violate the independence assumption required for standard ordinal logistic regression. To account for this structure, cumulative link mixed models (CLMM) were used, incorporating a random effect to model subject-specific variability. Several models were fitted using different sleep disturbance variables as response outcomes, including sleep quality, trouble sleeping, waking during the night, waking early, and restlessness. Predictor variables included menopausal status, age, household income, and hormone levels such as estradiol (E2) and follicle-stimulating hormone (FSH). Model selection was performed using Akaike’s Information Criterion (AIC) to identify the models that provided the best balance between fit and complexity. The results suggest that sleep disturbances become more common as women progress through the menopausal transition, with the highest likelihood of disturbances occurring during the transition from early to late peri-menopause. Higher income levels were generally associated with lower odds of frequent sleep disturbances, while hormonal changes were also related to sleep outcomes. In particular, higher estradiol levels were associated with better sleep outcomes, while increases in FSH were linked to greater odds of sleep disturbances. These findings highlight the combined influence of biological and socioeconomic factors on sleep health during midlife and demonstrate the usefulness of cumulative link mixed models for analyzing longitudinal ordinal health outcomes.

Advisor

Kelvey, Robert

Department

Mathematics

Disciplines

Mathematics | Physical Sciences and Mathematics | Public Health

Keywords

Women, Midlife, Menopause, Hormones, Sleep Health, Sleep

Publication Date

2026

Degree Granted

Bachelor of Arts

Document Type

Senior Independent Study Thesis

Download

Share

COinS
 

© Copyright 2026 Brandy Kaminta