Modeling of Bounded Outcome Scores with Data on the Boundaries: Application to Disability Assessment for Dementia Scores in Alzheimer’s Disease

摘要: Mixed-effects beta regression (BR), boundary-inflated (ZOI), and coarsening model (CO) were investigated for analyzing bounded outcome scores with data at the boundaries in context of Alzheimer’s disease. Monte Carlo simulations conducted to simulate disability assessment dementia (DAD) using these three models, each set simulated analyzed by original simulation model. One thousand trials simulated, trial contained 250 subjects. For subject, DAD baseline, 13, 26, 39, 52, 65, 78 weeks. The simulation-reestimation exercise showed that all models could reasonably recover their true parameter values. bias estimates ZOI was generally less than 1%, while CO mainly within 5%. BR slightly higher, i.e., or order 20%. In application real-world from clinical studies, examination prediction error visual predictive check (VPC) plots suggested both had similar performance described longitudinal progression better conclusion, modeling approaches may be sensible choices on edges. Prediction VPC can used identify best performance.