Statistical Challenges and Analysis of Longitudinal Mini-Mental State Examination Data in Frontotemporal Lobar Degeneration and Alzheimer's Disease

• Activity Number: 424
• Type: Contributed
• Date/Time: Tuesday, July 31, 2012 : 2:00 PM to 3:50 PM
• Sponsor: Statistics in Mental Health Research
• Author(s): Kay-See Tan*+ and David J. Libon and Katya Rascovsky and Murray Grossman and Sharon Xie
• Companies: University of Pennsylvania and Drexel University College of Medicine and University of Pennsylvania and University of Pennsylvania and University of Pennsylvania
• Keywords: MMSE ; Alzheimer's disease ; frontotemporal lobe dementia ; longitudinal assessment ; mixed-effects models ; cognitive decline

Abstract

Cognitive decline is of great interest in neurodegenerative disease research. Floor and ceiling effects as well as missing data are potential challenges in analyzing longitudinal cognitive data. In this paper, we compare several statistical approaches such as mixed-effects models and piecewise regression for analyzing a common dementia severity measure, Mini-Mental State Examination (MMSE), over time. Data from 213 patients diagnosed with Alzheimer's disease (AD) and four frontotemporal lobar degeneration (FTLD) phenotypes (behavioral variant frontotemporal dementia [bvFTD], nonfluent agrammatic variant of primary progressive aphasia [nfaPPA], semantic variant PPA [svPPA], and corticobasal syndrome [CBS]) were collected over five years. Results indicated MMSE declined faster for AD, svPPA and CBS compared to bvFTD and nfaPPA patients. Based on comparisons between MMSE profiles and clinical understanding of these different forms of dementia, we conclude that the MMSE may not reflect cognitive decline equally in all forms of dementia, hence necessitating neuropsychological instrument that can assess the longitudinal patterns of cognition decline specific to AD and FTLD subtypes.