Abstract

In this paper, we propose a method for jointly analysing survival data and longitudinally measured ordinal data and survival data with competing risks. A cause-specific sub-model is used to model competing risks survival data and a proportional odds mixed model is used for the ordinal longitudinal outcomes, where these outcome processes are linked through a set of association parameters. A primary objective of this study is to examine the effects of association parameters on the estimators of joint models. We apply the method to the Heart Outcomes Prevention Evaluation – 3 (HOPE-3) trial. We consider systolic blood as a longitudinal ordinal outcome. Both systolic blood pressure and diastolic blood pressure could be treated as multivariate longitudinal ordinal outcomes.