In the joint modeling, various outcomes are connected by a set of association parameters via random effects. We examine the influence of association parameters on the estimators of the joint and separate model parameters. We studied joint modeling of three outcomes: continuous longitudinal variable, ordinal longitudinal variable, and time-to-event outcome variable. We use a random effects accelerated failure time model for time-to-event outcomes, a linear mixed model for continuous longitudinal outcomes and a proportional odds mixed model for ordinal longitudinal outcomes, where these response models are linked through a set of association parameters. The proposed joint model parameters are estimated by the method of maximum likelihood. The finite-sample properties of the estimators are studied using Monte Carlo simulations. In clinical sciences, designing a study with multiple endpoints is a common phenomenon while joint analysis of these endpoints is a rare phenomenon. Thus we applied our proposed methodology to a clinical study, referred to as the Genetic and Inflammatory Markers of Sepsis (GenIMS) study.