Objective: Cluster analysis based on Alzheimer's disease (AD) neuropsychiatric profile demonstrated validity on caregiver burden, nursing-home placement and survival. The aims of our study were to explore the validity of this approach on caregiver burden, lack of insight and cognitive impairment and to examine the impact of neuropsychiatric profiles on these variables.
Method: A data-driven approach (two-step cluster analysis) identified groups of patients based on similarities of their neuropsychiatric symptom profile, as assessed by the Neuropsychiatric Inventory (NPI). ANOVAs and chi(2) tests were used to compare groups with regard to continuous and categorical variables. Linear regressions tested the relationships between NPI and clinical variables.
Results: Psychotic/behavioral, depressive and minimally symptomatic clusters differed for caregiver burden and lack of insight. Patients in the minimally symptomatic cluster showed better insight than those in the depressive cluster. Caregivers of the psychotic/behavioral cluster experienced the highest burden. We found positive relationships between NPI and lack of insight in the depressive and minimally symptomatic clusters and between NPI and caregiver burden in all three clusters. Caregiver burden was influenced by the type of symptoms.
Conclusions: The cluster analysis was valid for lack of insight and caregiver burden. Symptoms predominant on caregiver burden could become targets for therapy.
Copyright 2010 S. Karger AG, Basel.