of this scale in MCI patients both in Brazil and worldwide. Methods: 55 older adults with MCI (Petersen’s criteria) were assessed using the COPM. Test-retest was accessed in 27 MCI patients. An occupational therapist made the COPM interview and repeated the test 2 weeks later. The reliability of inter-rater agreement was accessed in the remaining 28 patients, which were interviewed with COPM by two different occupa- tional therapists in the same day . Pearson’s correlation test was used in the analysis. Results: Correlation of the problems identified (ex.: climbing stairs, paying bills, etc) in the performance of daily activities in the test- retest was 0.74 (p < 0.001) and for inter-rater agreement was 0.72 (p < 0.001). For the problems identified a score was given for the level of per- formance. In the test-retest the correlation of this score was 0.09 (p ¼ 0.776) and in the inter-rater evaluation it was 0.66 (p < 0.001). A score of satisfaction with the performance is also given through the COPM and in the test-retest the correlation was -0.11 (p ¼ 0.715) and in the in- ter-rater it was 0.67 (p < 0.001). Conclusions: Results showed that the COPM has generally a good reliability both in the test-retest and in the in- ter-rater agreement for problems identified. Less consistent results were found to the performance perceived and level of satisfaction with the prob- lems in the test-retest, perhaps because it depends on more subjective pa- rameters and a longer period (2 weeks) has passed between interviews. The COPM can be used to evaluate problems in the performance of daily activities in MCI elderly patients. P1-143 THE MONTREAL COGNITIVE ASSESSMENT (MOCA): VALIDATION OF ALTERNATE FORMS AND NEW RECOMMENDATIONS FOR EDUCATION CORRECTIONS Howard Chertkow 1 , Ziad Nasreddine 1 , Erin Johns 2 , Natalie Phillips 3 , Cristina McHenry 2 , 1 McGill University, Montreal, Quebec; 2 Concordia University, Montreal, Quebec; 3 Concordia University, Montreal, Quebec. Background: The Montreal Cognitive Assessment (MoCA) is a popular cognitive screening test designed to detect mild cognitive impairment (MCI) in older adults (Nasreddine et al., 2005). It was originally validated in a normative sample with a mean education of 13.3 years using a cutoff score of 26. Our objective was to obtain normative data in a sample with 12 years or less of education, and to develop alternate forms to facilitate its use in repeated testing. Methods: Ninety-six healthy participants without cognitive problems were tested on the original MoCA and a neuropsychological battery. This sample was combined with lower-education participants from the initial validation study. Fifty-five participants were excluded, mainly due to poor neuropsychological performance (score ¼ 1.5 SD below the mean on 1+ measure). The final sample was 79participants. Two alternate MoCA forms were developed by replacing original items with new exemplars. The original and two new forms were administered one month apart in randomized order to 32normal elderly controls, 30 patients with MCI, and 21 patients with Alz- heimer disease. Results: 1. Revised education corrections of +1 point for 10- 12 years of education (n ¼ 52) and +2 points for4-9 years of education (n ¼ 27) are suggested. Sensitivity for detecting MCI/AD were 90/100% and speci- ficity 69.2%for 10-12 years of education, and 87.5/100% sensitivity, 74.1% specificity for4-9 years of education. 2. Repeated measures ANOVA on the three MoCA forms revealed the following: total scores for the NECs and AD’s on all three versions were within 0.8 points of each other, with no signif- icant differences between the versions. Total scores for the MCIs showed a sig- nificant but small 1.3point difference between MoCA 1 and MoCA3. Conclusions: The MoCA has lower specificity for detecting MCI in lower ed- ucation samples, highlighting the challenge of cognitive screening in older adults with lower education but, overall, retains its excellent psychometric properties and sensitivity as a screening tool for MCI and mild AD. The three forms of the MoCA yield equivalent total scores and discriminate MCI patients from controls and AD patients. The three forms are suitable for situations re- quiring repeated cognitive testing P1-144 INTEGRATION OF COGNITIVE IMPAIRMENTS AND WHITE MATTER HYPERINTENSITIES INTO A 12-CELL DIAGNOSTIC MATRIX SYSTEM: ANALYSIS OF 3,966 SUBJECTS FROM THE CREDOS STUDY Chan-Seung Chung 1 , Joongsun Lee 2 , Hae-Kwan Cheong 3 , Kyung Ryeol Cha 4 , SangYun Kim 5 , Chang Hyung Hong 6 , Sang Won Seo 7 , Byeong Kil Yeon 8 , Duk Na 9 , Byoung Hoon Oh 10 , Doh Kwan Kim 11 , Han Seol-Heui 12 , Seong Yoon Kim 13 , Jae-Hong Lee 14 , 1 Gyeonggi Provincial Yongin Geriatric Hospital, Yongin-si; 2 Asan Medical Center, Seoul; 3 Sungkyunkwan University School of Medicine, Suwon; 4 Mungyeong Jeil General Hospital, Mungyeong si; 5 Seoul National University Bundang Hospital, Seoul National University College of Medicine, Seongnam-si; 6 Ajou University School of Medicine, Suwon; 7 Samsung Medical Center, Seoul; 8 Kangdong Sacred Heart Hospital, Seoul; 9 Samsung Medical Center, Seoul; 10 School of Medicine, Yonsei University, Seoul; 11 Samsung Medical Center, Seoul; 12 Konkuk University School of Medicine, Seoul; 13 Asan Medical Center, Seoul; 14 Asan Medical Center, Seoul. Background: Mild cognitive impairment (MCI), especially amnestic type (aMCI), is known to be a preclinical stage of Alzheimer’s disease (AD), as is subcortical vascular MCI (svMCI) a prodromal stage of subcortical vas- cular dementia (SVaD). Clinically, it is difficult to draw a line between AD versus SVaD and between aMCI versus svMCI. Rather, these conditions should be viewed as being in the same spectrum with AD or aMCI on one end of the continuum and SVaD or svMCI on the other end of the con- tinuum. The goal of this study was to introduce a new method of integrat- ing these spectrum disorders into so-called “12-cell Diagnostic Matrix” and to report the demographic and clinical characteristics of subjects from each cell. Methods: The Clinical Research for Dementia of South Korea (CREDOS) recruited 4,228 subjects between 2005 and January 30, 2010 from 56 hospitals, who underwent the identical evaluation pro- tocol including neuropsychological tests and MRI. Subjects were to have one of the five diagnostic categories: no cognitive impairment (NCI), aMCI, svMCI, AD, or SVaD. However, rather than the AD/SVaD or aMCI/svMCI dichotomy, we rearranged the subjects into a 12 cell ma- trix that consisted of four cognition levels (NCI, MCI, mild to moderate dementia, and severe dementia) and three white matter ischemia burdens on MRI (minimal, moderate, and severe)(figure 1). Results: About half of the subjects (N ¼ 2,114; 50.0%) had mild to moderate dementia, of which the ischemia severity distribution were 49.7:34.1:16.2 (%, mild: moder- ate: severe). For patients with MCI (N ¼ 1,566; 37.0%), the ischemia se- verity distribution was 65.5%, 26.8%, and 7.7%, respectively. The mean age was 71.9 6 8.2 years. There was a female-to-male ratio 7:3 and an av- erage of 7.0 6 5.3 years of education (table 1). Conclusions: The CRE- DOS study which consisted of subjects with normal cognition, patients with aMCI versus svMCI, AD versus SVaD, and those within the grey zone between them will provide a comprehensive perspective of MCI and dementia and serve as a model for future cross-sectional and longitu- dinal studies. P1-145 A STUDY ON THE CLINICAL PROFILE AND SOCIO-CULTURAL FACTORS INFLUENCING DIAGNOSIS OF MILD COGNITIVE IMPAIRMENT AND DEMENTIA FROM A MEMORY CLINIC IN URBAN SOUTH INDIA Ratnavalli Ellajosyula 1 , Sushil Ingole 2 , Rupa Ananthasivan 3 , 1 Manipal Hospital; 2 Manipal hospital, Bangalore; 3 Manipal hospital, Banglaore. Background: Characterization of mild cognitive impairment (MCI) and dementia subtypes has become increasingly important for treat- ment and prognosis. The geographical location and socio-cultural Poster Presentations P1 S157