How Well Do the ADAS-cog and its Subscales Measure Cognitive Dysfunction in Alzheimer’s Disease?

Abstract

Background/Aims: The Alzheimer’s Disease Assessment Scale-cognitive (ADAS-cog) is regularly used to assess cognitive dysfunction in Alzheimer’s disease (AD) clinical trials. Yet, little is known about how the instrument and its subscales measure cognition across the spectrum of AD. The current investigation used item response theory (IRT) analyses to assess the measurement properties of the ADAS-cog across the range of cognitive dysfunction in AD. Methods: We used IRT-based analyses to establish the relationship between cognitive dysfunction and the probability of obtaining observed scores on each subscale and the test as a whole. Data were obtained from 1,087 patients with AD and amnestic mild cognitive impairment. Results: Results showed that the ADAS-cog and its subscales provide maximum information at moderate levels of cognitive dysfunction. Raw score differences toward the lower and higher ends of the scale corresponded to large differences in cognitive dysfunction, whereas raw score differences toward the middle of the scale corresponded to smaller differences. Conclusions: The utility of the ADAS-cog and its subscales is optimal in the moderate range of cognitive dysfunction, but raw score differences in that region correspond to relatively small differences in cognitive dysfunction. Implications for tracking and staging dementia and for clinical trials are discussed.