Clinical Manifestations

Area of Science:

• Neuroscience
• Psychometrics
• Clinical Trials

Background:

• Cognitive impairment is central to Alzheimer's disease (AD), necessitating reliable assessment tools for research and treatment.
• The Alzheimer's Disease Assessment Scale-Cognitive Subscale (ADAS-Cog) is the standard measure in AD clinical trials.
• Traditional psychometric methods have been supplemented by Item Response Theory (IRT) for more nuanced evaluation of assessment tools.

Purpose of the Study:

• To utilize Item Response Theory (IRT) to identify specific tasks within the ADAS-Cog that effectively discriminate varying levels of cognitive impairment in Alzheimer's disease patients.
• To analyze individual-level data from completed clinical trials to refine the understanding of ADAS-Cog item performance.

Main Methods:

• Employed the Graded Response Model, a form of IRT, analyzing data from 2,198 participants across five randomized controlled trials of donepezil.
• Calculated item discrimination parameters (α) to assess the ability of each ADAS-Cog item to differentiate cognitive impairment severity.
• Defined thresholds for item discrimination: very low (0.01–0.24), low (0.25–0.64), moderate (0.65–1.34), high (1.35–1.69), and very high (>1.7).

Main Results:

• Word recall demonstrated the highest discrimination ability (α=1.92).
• Four items (Spoken language ability, Comprehension of spoken language, Remembering test instruction, Word finding difficulty) exhibited low discrimination, indicating limited ability to distinguish impairment levels.
• Seven items (Word recall, Commands, Naming, Constructional praxis, Ideation praxis, Orientation, Word recognition) were identified as effective discriminators of cognitive impairment.

Conclusions:

• Seven ADAS-Cog items effectively differentiated between placebo and donepezil treatment groups in pivotal Alzheimer's disease trials.
• This study enhances understanding of AD assessment by pinpointing tasks crucial for monitoring cognitive function, particularly in moderate-stage Alzheimer's disease.
• The identified discriminating tasks offer potential for improved treatment monitoring in clinical practice and future research.