Joint modeling of cross-sectional health outcomes and longitudinal predictors via mixtures of means and variances

• Published in: Biometrics Vol. 71; no. 2; pp. 487 - 497
• Main Authors: Jiang, Bei, Elliott, Michael R., Sammel, Mary D., Wang, Naisyin
• Format: Journal Article
• Language: English
• Published: United States Blackwell Publishing Ltd 01.06.2015; International Biometric Society
• Subjects: Area Under Curve; BIOMETRIC PRACTICE; Biometrics; Biometry; Computer Simulation; Cross-Sectional Studies; Female; Follicle Stimulating Hormone, Human - metabolism; Hot Flashes - etiology; Hot Flashes - metabolism; Humans; Joint model; Latent class; Long-term trend; Longitudinal Studies; Menopause; Model misspecification; Models, Statistical; Outcome Assessment (Health Care) - statistics & numerical data; Predictive performance; Regression Analysis; Risk Factors; Shared random effects and variances; Short-term variability; Simulation; Womens health; Latent class; Short-term variability; Long-term trend; Shared random effects and variances; Joint model; Model misspecification; Predictive performance
• ISSN: 0006-341X; 1541-0420
• DOI: 10.1111/biom.12284

Joint modeling methods have become popular tools to link important features extracted from longitudinal data to a primary event. While most modeling strategies have focused on the association between the longitudinal mean trajectories and risk of an event, we consider joint models that incorporate information from both long-term trends and short-term variability in a longitudinal submodel. We also consider both shared random effect and latent class (LC) approaches in the primary-outcome model to predict a binary outcome of interest. We develop simulation studies to compare and contrast these two modeling strategies; in particular, we study in detail the effects of the primary-outcome model misspecification. Among other findings, we note that when we analyze data from a shared random-effect using a LC model while the information from the longitudinal data is weak, the LC approach is more sensitive to such a model misspecification. Under this setting, the LC model has a superior performance in within-sample prediction that cannot be duplicated when predicting new samples. This is a unique feature of the LC approach that is new as far as we know to the existing literature. Finally, we use the proposed models to study how follicle stimulating hormone (FSH) trajectories are related to the risk of developing severe hot flashes for participating women in the Penn Ovarian Aging Study.