Introduction: Dependence in Alzheimer disease has been proposed as a holistic, transparent, and meaningful representation of disease severity. Modeling clusters in dependence trajectories can help understand changes in disease course and care cost over time.
Methods: Sample consisted of 199 initially community-living patients with probable Alzheimer disease recruited from 3 academic medical centers in the United States followed for up to 10 years and had ≥2 Dependence Scale recorded. Nonparametric K-means cluster analysis for longitudinal data (KmL) was used to identify dependence clusters. Medicare expenditures data (1999-2010) were compared between clusters.
Results: KmL identified 2 distinct Dependence Scale clusters: (A) high initial dependence, faster decline, and (B) low initial dependence, slower decline. Adjusting for patient characteristics, 6-month Medicare expenditures increased over time with widening between-cluster differences.
Discussion: Dependence captures dementia care costs over time. Better characterization of dependence clusters has significant implications for understanding disease progression, trial design and care planning.