Introduction Heterogeneity is observed in the patterns of cognition in Alzheimer’s disease (AD). 4 bad status. 4) genotype in predicting class regular membership and in exploring the part of further covariates after adjustment for these factors. Method The analysis dataset of AD cases was drawn from a large case-control study of 875 AD individuals and 850 non-demented control subjects recruited from nine Memory space Referral Clinics in Canada between 6/2002 and 3/2005 explained elsewhere (Li et al., 2008). The study protocol included neurological, neuropsychological and laboratory assessments plus medical record review of dementia history (including neuroimaging) where available. Inclusion criteria required that AD patients fulfilled criteria defined in DSM-IV (American Psychiatric Association, 1994)and by NINCDS-ADRDA (McKhann et al., 1984) criteria for probable AD, with a Global Deterioration Level (GDS) of 3-7 (ranging from slight to very severe cognitive decrease) (Reisberg B et al., 1982) Subjects were excluded if they were in a major depressive episode, acute psychosis, or acute manic or depressive episode of bipolar disorder at the time of recruitment. Neuroimaging was not required as part of the study protocol although imaging at BILN 2061 the time of AD diagnosis to rule out vascular and other causes of dementia would have been expected clinical practice. The study protocol was examined and authorized by the appropriate ethics committee (EC) or investigational review table (IRB) for each study site prior to subject recruitment. BILN 2061 Informed consent was from study participants in accordance with all relevant IRB/EC and regulatory requirements. The present study sample was restricted to 627 slight/moderate AD cases based on a total Mini-Mental Status Exam (MMSE) (Folstein et al., 1975) score of 15, to limit the influence of floor effects within the cognitive scales in severe AD. Cognitive Assessment Cognitive function was assessed with the MMSE (Folstein et BILN 2061 al., 1975) and the Mattis Dementia BILN 2061 Rating Level-2 (DRS-2) (Mattis, 1976; Jurica et al., 2001) scales. Scores on a total of 11 subscales from these checks were used in LCA to derive subgroups of cognitively related patients based on impairment in specific cognitive BILN 2061 domains. The DRS-2 subscales were defined relating to Jurica et al (2001) as Attention, Conceptualization, Building, Initiation/Perseveration and Memory. The MMSE questions were grouped into the following categories: attention (spell WORLD backwards), language (object naming, phrase repetition, writing a phrase, read and follow control Close your eyes), orientation (for time and place), memory space (sign up and recall of apple, penny, table), praxis (3-stage control) and building (pentagon copy). Data Analysis Latent class analysis (LCA) of the 11 cognition subscale items was used to examine the latent structure of cognition in the sample of AD cases. In order to right both for variations in range of possible scores on each subscale (1-37) which might affect weighting of the variable in the analysis, and for skewness of subscale score distributions, median total sample scores for each subscale were used as slice points to produce dichotomous indicators for each subscale, related to high/low scores based on the sample distribution. Low scores on both the MMSE and DRS-2 indicate higher impairment. LCA is definitely a probability-based clustering method which assumes that associations between individuals, based on reactions MADH9 for the observed items, can be explained by an underlying class structure (McCutcheon, 1987). This structure can be characterised through observation of the structural model consisting of latent class probabilities (guidelines which correspond to latent class prevalence) and the measurement model or item response probabilities, conditional on class membership (guidelines) (Lanza S.T. et al., 2007a). In the current analysis, the guidelines will correspond to the prevalence of each cognition class and the guidelines, corresponding to probability of low scores in each of the measured cognitive domains, can be used to infer the cognitive profile associated with class membership. It is assumed that within each latent class, individual items will be.