Title: Latent class growth analysis for ordinal response data in the Distress Assessment and Response Tool: an evaluation of state-of-the-art implementations

Abstract: Latent class growth analysis is a popular approach to identify underlying subpopulations. Several implementations, such as LCGA (Mplus), Proc Traj (SAS) and lcmm (R) are specially designed for this purpose. Motivated by data collection of psychological instruments over time in a large North American cancer centre, we compare these implementations using various simulated Edmonton Symptom Assessment System revised (ESAS-r) scores, an ordinal outcome from 0 to 10, as well as the real data consisting of more than 20,000 patients. We found that Mplus and lcmm lead to high correct classification rate, but Proc Traj over estimated the number of classes and failed to converge. While Mplus is computationally faster than lcmm, it does not allow more than 10 levels. We therefore suggest first analyzing data on the ordinal scale using lcmm. If computational time becomes an issue, then one can group the scores into categories and implement them in Mplus.