It has become increasingly appreciated that practice effects and ceiling effects could result in type 1 or type 2 errors in clinical trials involving early stage AD populations, given the necessity of serial assessments. This proposal has the overarching goal of validating novel cognitive and everyday functional measures that have sharply attenuated practice effects and are not prone to ceiling effects for use in preclinical Alzheimer's disease (AD) trials in which participants are cognitively within normal limits. To implement this, we will conduct an innovative parallel group study in which 320 healthy, non-cognitively impaired older subjects are randomized to one of two groups based on assessment type (novel instruments vs. established) and receive three serial assessments over a one year period. We will employ this parallel group design in order to maintain the structure of a clinical trial, while pari passu, highlighting contrasts between novel and established measures and completely avoiding any interference effects between them. Our novel cognitive measures include tests of executive function, episodic memory, and processing speed combined into a single composite. Our functional measures involve computerized performance based, ecologically relevant instrumental activities. We will compare our novel No Practice Effects (NPE) cognitive battery and Miami Computerized Functional Assessment Scale (CFAS) against established measures that include the PACC, ADAS-Cog, and FAQ. In preliminary data our novel measures demonstrated attenuated practice effects, high test-retest reliabilities, equivalent alternate forms, and lack of ceiling effects. Moreover, we will also examine several biomarkers and markers that may increase risk for AD (e.g., APOE e4 carrier status, hippocampal atrophy and AD cortical thinning signature, family history of AD) in order to assess the sensitivity of the measures to these known biomarkers and so provide evidence for or against the validity of the cognitive and functional tests.
In this new Administrative Supplement we will design versions of the novel tests that can be administered remotely by telephone or smartphone. These include Recognition Memory, Digit Symbol, Verbal Fluency, N Back, Brown Peterson, and Letter Number Span, as well as a measure of everyday functional competence (prescription refill). We will empirically assess the remotely delivered versions have correspondence to the original paper and pencil versions.
