A Hierarchical Integration of Person-Centered Comorbidity Models: Structure, Stability, and Transition Over Time

Person-centered analyses of mental disorder comorbidity typically identify a single optimal latent group structure. We took an alternate approach, modeling every estimable comorbidity class model in a nationally representative sample (N = 34,653) and integrating them into a single overarching hierarchy, providing a full multilevel accounting of various person-centered comorbidity structures. We then investigated the structural stability of this hierarchy across two waves of data collection, and the stability of, and transition between, comorbidity classes over time using latent transition analysis. Findings suggested that comorbidity classes were structured into an interpretable hierarchy. Evidence for robust structural stability of the hierarchy over time was found, regardless of assessment time points and diagnostic time frames used. Latent transition analysis provided evidence for both continuity and change of comorbidity group membership. We discuss how person- and variable-centered models provide complementary perspectives toward empirically supported classification.