In this modeling approach, DTSA is a special case within the general latent variable framework which corresponds to a single-class latent class analysis with binary time-specific event indicators. The first step was to fit an unconditional survival model that included only the five binary time-specific event indicators for the remission of AUDs across adulthood. The constant hazard assumption was then evaluated by comparing the unconditional survival model, which allowed the hazard rate to vary across time, to a model that constrained the hazard rate to equality across intervals using a likelihood-ratio test (LRT) based on the model deviance statistics. The next step was to evaluate if predictors related to the outcome in a similar manner across ages regarding each predictor, and by comparing a model with time-varying effects to a model that constrains the predictor/covariate effects to being equal over time using a series of LRTs. If a characteristic violated proportionality, it was used in the DTSA in a manner that allows for the time-varying effects. Finally, a multivariate model that included all predictors was evaluated to allow interpretation of how each individual characteristic related to the outcome of after adjusting for all other predictors.
