A joint modeling approach to study the association between subject-level longitudinal marker variabilities and repeated outcomes

Irena Chen, Zhenke Wu, Sioban D Harlow, Carrie A Karvonen-Gutierrez, Michelle M Hood, Michael R Elliott (2023+). Submitted.
Full text


Women experience increased risk of bone loss during the menopausal transition (MT); in fact, nearly 50% of women’s lifetime bone loss occurs during this time. In addition to mean levels, variability in certain hormones may be an important predictor of bone health. We introduce a joint model that characterizes individual mean trajectories and the individual residual variances and links these variances to bone health trajectories. Higher estradiol variability was associated with faster decreases in bone mineral density (BMD) during MT. Higher follicle-stimulating hormone variability was associated with larger declines in BMD around menopause, but this association was moderated over time.

Keywords Bone mineral density, Estradiol, Follicle-stimulating hormone, Joint models; Subject-level variability; Variance component priors