Multi-wave studies, particularly those employing ecological momentary assessments (EMAs), provide a rich opportunity to capture nuanced patterns of dynamic changes in individuals' behaviors and psychological states over time. Understanding both within-wave (e.g., daily fluctuations) and between-wave (e.g., pre-, during, and post-intervention periods) longitudinal changes is critical for identifying intervention effects not only on mean/baseline levels but also on dynamic features such as intraindividual variability and temporal correlations between observations. Existing analytical approaches often do not effectively capitalize on known (e.g., study design-related) information to disentangle meaningful between- and within-wave changes, while simultaneously accommodating individual differences in these changes. In this talk, a novel Bayesian modeling framework designed to simultaneously capture within- and between-wave longitudinal changes will be introduced. This approach incorporates a multilevel structure to account for individual differences in these changes, offering a more tailored understanding of intervention effects. The approach is examined via simulation studies and applied to a mobile health intervention study with EMA data collected from different intervention periods. Several key issues related to detecting multi-wave changes and potential solutions will be discussed.

REGISTER