Temporal quantification of Alzheimer's disease severity: 'time index' model

Abstract

A fundamental issue in the clinical and neuropathological assessment of Alzheimer's disease patients is quantification of dementia severity progression. Several methods have been advanced for the purpose of staging dementia with various sensitivities at different phases of the disease, but no mathematical function has been developed to link these measures to a physical continuum. Using a dynamic method for quantifying illness severity, change in severity over time was referenced to a cumulative temporal index, a physical dimension. Data from 33 patients with probable Alzheimer's disease with at least 2 successive assessments on three 50-point scales measuring cognitive, behavioral, and daily living skills were used to determine rate of change. 'Fuzzylogic' smoothing of the data, integration over time, and least-squares regression were used to derive a cubic polynomial function to calculate a severity measure in which 'days of illness' was estimated from the severity score. This method can be used to improve the comparability of performance across various mental status tests, and to link measures of very early phases of preclinical dementia and late profound dementia phases. This method also provides a description of an 'average' time course for any population from which the index is derived.