Keywords: multi-state models
Multimorbidity is the occurrence of two or more chronic diseases. Multiple approaches have been used to identify patterns of multimorbidity, including focusing on dyads and triads of conditions, clustering, factor analysis and supervised learning. However, most investigations have utilized cross-sectional data, which has limited the information obtained. Longitudinal data on multimorbidity can be examined using multi-state models. In this context, each state is an individual morbidity or a combination of morbidities. These models provide information on the prevalence of the cohort in each state, the probability of transitioning to another state and the time spent in each state. Graphical approaches to displaying these results will also be discussed.