Tonmoy Sharma

P4-354: Cognitive performance in healthy volunteers on the Cogtest computerized neurocognitive battery across cultures: Implications for clinical trials of Alzheimer's disease

relateddisability and long term care placement, there remarkably little large clinical trial experience in the primary prevention of dementia. We report participant adherence and retention data from the Ginkgo Evaluation of Memory Study (GEMS), a trial of the effectiveness of Ginkgo biloba in delaying the incidence of AD/all cause dementia. Objectives: The purpose of this study was to assess the utility of Ginkgo biloba 120 mg BID on delaying emergence of AD/all cause dementia. We herein provide data on adherence and retention that will be useful in planning and carrying out clinical trials in older and minority populations. Methods: A randomized, double-blind, placebo-controlled clinical trial conducted at 4 academic sites in the US. A total of 3,071 subjects age 75 with normal cognition or MCI were randomized to either Ginkgo or placebo after extensive medical and neuropsychological screening, including the Clinical Dementia Rating (CDR) performed with each participant’s proxy. Assessments, including medication compliance, were performed every 6 months and incorporated home visits, proxy interviews and telephone assessments. Results: A total of 440 persons developed dementia over the course of the study, extending 8 years from initial recruiting to last peron/last visit. At trial close out cognitive data and vital status were available on a very high percentage of participants. We will present the methods incorporated to maximize adherence and retention as well as visit and medication compliance results in this older population (Mean age 83 at end of trial). Conclusions: Dementia and cognition are assessable outcomes in primary prevention trials and adherence and retention can be maintained in at-risk elderly cohorts. Substantial information from GEMS will be of benefit to designing future prevention trials assessing medication effectiveness, and the potential incorporation of retention/compliance behaviors as an outcome or outcome modifier.

P1-154: Cogtest identifies impairment in multiple memory domains in age-associated memory impairment

Introduction Age Associated Memory Impairment (AAMI) has been used as an objective way of defining cognitive decline with ageing. The concept, developed by Crook et al. in 19861, has well-defined criteria. AAMI was originally developed to refer to a subpopulation of normal older individuals1, but there is evidence to suggest that it may be a discrete entity, showing significant brain and behavioral changes compared to normal ageing2,3,4. The purpose of this study was to examine the extent to which memory is affected in AAMI using COGTEST. Auditory Working Memory: AAMI group sequenced fewer numbers than the control group, however this did not represent a significant group difference.

P4-359: Memory deficits in mild cognitive impairment are identified with cogtest

Background: Understanding the memory impairment associated with mild cognitive impairment (MCI) and identifying assessment tools capable of measuring these impairments continues to be of paramount importance. The goal of this study was to examine the utility of Cogtest, a computerized neuropsychological test battery used with a variety of clinical populations and in clinical trials, to identify memory deficits in MCI. Methods: Cogtest (www.cogtest.com) has a library of computerized cognitive assessments with a platform allowing for accurate recording of reaction times and enhanced standardization of administration relative to conventional paperpencil tests The Word List Memory Test (verbal selective reminding memory test), Face Memory Test (visual memory), and the Auditory Number Sequencing Test (working memory) were administered to MCI (n 44) and healthy controls (HC, n 50). All subjects received a screening battery which included a paired associate test. A z score -1.5 SD below that of a normative group on this test was used as the objective measure of memory impairment in the definition of MCI. Results: Group differences were found for education and this was covaried in all analyses of Cogtest variables. ANCOVA were used to examine Cogtest variables and significance levels were set to p .01. Significant Group performance differences were seen on the Word List Memory Test (a computerized version of the Buschke selective reminding test). MCI subjects remembered significantly fewer words on the first trial (mean 4.8), on all trials (mean 36.3), and after a 30 minute delay (mean 5.8) compared to HC (mean’s 9.2, 54.3, 12.8, respectively). In the Face Memory Test, MCI subjects remembered significantly fewer words compared to HC (62% verses 78%, p .001. After a 30 minute delay, MCI subjects remembered significantly fewer faces (63%) compared to HC (73%) but not from each other. In the Auditory Number Sequencing task, MCI subjects sequenced significantly fewer numbers (mean 9.9) compared to HC (mean 12.2). Conclusions: We conclude that multiple memory domains as assessed with Cogtest are affected in MCI. The Cogtest library of tests is able to identify cognitive deficits in MCI patients.