UNCORRECTED PROOF The factorial structure Q1 of the mini mental state examination (MMSE) in Alzheimer’s disease A. Brugnolo a, *, F. Nobili a , M.P. Barbieri b , B. Dessi a , A. Ferro c , N. Girtler d , E. Palummeri b , D. Partinico e , U. Raiteri f , G. Regesta f , G. Servetto c , P. Tanganelli e , V. Uva g , D. Mazzei a , S. Donadio h , F. De Carli i , G. Colazzo j , C. Serrati j , G. Rodriguez a a Clinical Neurophysiology Unit, Department of Endocrinological and Medical Science, University of Genoa, Viale Benedetto XV 6, I-16132 Genoa, Italy b Department of Q2 Geriatrics, Galliera Hospital, Mura delle Capuccine 14, I-16142 Genoa, Italy c Department of Psychiatry, ASL 2, Via Manzoni 14, I-17100 Savona, Italy d Clinical Psychology Unit, Department of Law Medicine, University of Genoa, Largo G. Gaslini 5, I-16148 Genoa, Italy e Neurology Unit, Padre Antero Hospital, Sestri Ponente, Via Domenico Oliva 22, Sestri Ponente, I-16153 Genoa, Italy f Department of Neurology, S. Martino Hospital, Largo R. Benzi 10, I-16132 Genoa, Italy g Department of Psychiatry, ASL 4, Via G.B. Ghio 3, Chiavari, I-16043 Genoa, Italy h Department of Physics, University of Genoa, Via Dodecaneso 33, I-16146 Genoa, Italy i Institute of Molecular Bioimaging and Physiology, National Research Council, Via de Toni 5, I-16132 Genoa, Italy j Neurology Department, ASL 1 General Hospital, Via Aurelia Ponente 97, Bussana di Sanremo, I-18038 Imperia, Italy 1. Introduction The MMSE (Folstein et al., 1975) is probably the most popular measure to screen for cognitive impairment. The 30 MMSE items were firstly identified to distinguish between neurological and psychiatric patients. The MMSE was shown to have both good test– retest reliability (0.80–0.95) (Folstein et al., 1975; O’Connor et al., 1989; Tombaugh and McIntyre, 1992) and acceptable sensitivity (0.86) and specificity (0.92) in what the authors define ‘organic mental disorders’ (O’Connor et al., 1989). Although its application beyond the screening phase was criticized (Feher et al., 1992), it continues to be widely used both to assess the effect of therapeutic agents on cognitive function (Tierney et al., 2000; Meyer et al., 2002; Berger et al., 2005) and in the follow-up (Meyer et al., 2002). During the last two decades, several studies have investigated the item organization of the MMSE. A part of these studies concerned large elderly populations, including both normal subjects and patients with cognitive deterioration of different origin (Jones and Gallo, 2000; Shyu and Yip, 2001; Noale et al., 2006; Schultz-Larsen et al., 2007a,b). Other authors have focused attention in several ‘‘organic brain syndromes’’ leading to cognitive deficit (Feher et al., 1992; Tombaugh and McIntyre, 1992; Tierney Archives of Gerontology and Geriatrics xxx (2008) xxx–xxx 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 ARTICLE INFO Article history: Received 28 February 2008 Received in revised form 15 July 2008 Accepted 17 July 2008 Keywords: MMSE AD Factor analysis Dementia in elderly ABSTRACT Our aim was to evaluate the factorial structure of the mini mental state examination (MMSE) in Alzheimer’s disease (AD). Five hundred and twenty-four consecutive outpatients at their first diagnostic work-up (age 78.02 Æ 6.07 years, education 6.62 Æ 3.48 years, mean MMSE score 20.23 Æ 4.89) (ÆS.D.) with probable AD (based on DSM-IV and NINCDS-ADRDA criteria) were enrolled in a multicenter, cross-sectional, regional-based study. For the purpose of the present study, the 11 subtests composing the MMSE and the global MMSE score (ranging from 10 to 29, included) were considered. Factor analysis with Varimax rotation method identified two factors that explained about the 85% of total variance. The first factor explained the 65% of variance and mainly included temporal orientation, delayed recall, attention/concentration, and constructional praxia. The second factor explained the 20% of variance and included reading a sentence, writing a sentence, naming, verbal repetition and immediate memory. The first factor was a reliable index of cognitive deterioration along the MMSE score interval between 29 and 10, whereas the second factor was not a suitable marker in this range. The two-factor structure of the MMSE in AD is shown in a large series of patients. The first factor expresses the ability to use new information and is related with working memory. The second factor is related with a more consolidated knowledge, namely verbal abilities, and is essentially useless in mild to moderate AD. ß 2008 Published by Elsevier Ireland Ltd. * Corresponding author. Tel.: +39 010 3537 778; fax: +39 010 5556 893. E-mail address: Andrea.Brugnolo@unige.it (A. Brugnolo). G Model AGG 1906 1–6 Please cite this article in press as: Brugnolo, A., et al., The factorial structure of the mini mental state examination (MMSE) in Alzheimer’s disease, Arch Gerontol Geriatr (2008), doi:10.1016/j.archger.2008.07.005 Contents lists available at ScienceDirect Archives of Gerontology and Geriatrics journal homepage: www.elsevier.com/locate/archger 0167-4943/$ – see front matter ß 2008 Published by Elsevier Ireland Ltd. doi:10.1016/j.archger.2008.07.005