Background: Although cognitive problems have been identified in people with multiple sclerosis (PwMS), few studies have investigated the long-term change in cognitive functioning. Objective: To identify trajectories of change in cognitive functioning for PwMS. Methods: Participants enrolled in the quality-of-life subgroup from the Comprehensive Longitudinal Investigation of Multiple Sclerosis at Brigham and Women’s Hospital (CLIMB) were eligible for our analysis. In 2006, participants in this group began to complete the Symbol Digit Modalities Test (SDMT) annually. Latent trajectory models were used to identify groups of participants with similar longitudinal change in SDMT scores. Linear and quadratic trajectory models were fit, and the models were compared. Latent trajectory models were also fit adjusting for baseline age and disease duration as well as using normalized SDMT scores. The groups identified across the approaches were compared. Results: We found that classes with higher-than-average baseline values improved, classes with average baseline values remained relatively constant, and classes with lower baseline values experienced cognitive worsening. Similar results were observed in the alternative latent trajectory models accounting for other variables. Conclusion: Our models show that subjects with higher SDMT scores at baseline showed improvement, while subjects with lower SDMT scores at baseline showed worsening. Baseline age and disease duration were also associated with SDMT performance.
Erratum to: Using latent trajectory analysis of residuals to detect response shift in general health among patients with multiple sclerosis
