di
M. Gallucci a,e, P. Antuono b, F. Ongaro c , P.L. Forloni d, D. Albani d, G.P. Amici a,e, C. Regini a

a ARGEI, Interdisciplinary Geriatric Research Association, Treviso, Italy; b The Medical College of Wisconsin, Milwaukee, USA; c Department of Statistics, University of Padova, Italy; d Department of Neuroscience, Mario Negri Institute for Pharmacological Research, Milano, Italy; e General Hospital, Treviso, Italy

Purpose of research: The aim of this study was to examine correlations between physical activity, socialization, reading and cognitive decline in a sample of 668 elderly people over the age of 70, living in Treviso, a typical town in north-east of Italy. This study is part of the "TREVISO LONGEVA" Targeted Health Research Project, supported by the Veneto Region and other bodies, realized by ARGeI, that has collected biological, medical, social, economic, demographic and quality of life data.

Materials and methods: The population over the age of seventy living in the municipality of Treviso was divided according to sex and age into eight groups (with four age groups: 70-79, 80-89, 90-99, 100 and over) within which a predetermined number of individuals was extracted, 250, 200, 200 persons respectively, half male and half female and all those aged 100 and over. Systematic sampling within each of the 6 layers related to the population under 100 years of age. The sample, involving a total of 668 people, was then weighted with reference to the population distribution. Data were collected through an interview and a blood sample, both performed at home. Cognitive decline was assessed using the MMSE. With regard to physical activity, the questionnaire asked if the elderly took walks and/or did gardening every day. Socialization was examined by asking if intervieweds regularly visited friends (a) or recreation and/social centres (b) and if they were involved in socially useful activities (c) or voluntary work (d). As far as cultural activities were concerned, the elderly were asked if they read and what they read (newspapers, novels, non-fiction). In the population examined, we assessed comorbidity by considering the 19 most prognostically significant diseases according to the ICD-9 classification by calculating the two comorbidity prognostic indices, DCI (Disease Count Index, a simple sum of present diseases) and CCI (Charlson Comorbidity Index, a weighted sum of present diseases and age).

Results: Sample mean age is 84.1±8.1 years, 53.4% female, mean MMSE score is 23.7±6.8. The 179 subjects who practice both physical activity types (walking and gardening) show a MMSE score of 26.7±3.1 while the 336 subjects who practice only one of the two physical activity types show a MMSE score of 26.3 ± 3.9 and the 153 subjects who do not practice any physical activity show a MMSE score of 21.4±6.9 (ANOVA p<0.0001). The 91 subjects who practice more than one of the described socializing activities (a, b, c, d) show a MMSE score of 27.0±2.7 while the 249 subjects who practice only one of the described socializing activities show a MMSE score of 26.6±3.5 and the 328 subjects who do not practice any socializing activitiy show a MMSE score of 23.7±6.2 (ANOVA p<0.0001). Moreover, 172 subjects read novels and non-fiction (MMSE 27.3±2.5), 356 subjects read only newspapers (MMSE 26.0±3.9), and 140 subjects do not read at all (MMSE 21.0±7.3). There are significant cognitive differences between all the three categories (ANOVA p<0.0001, multiple comparisons Tukey-Kramer p<0.05). Applying a multiple regression model, comorbidity as CCI (p<0.0001), physical activity (p<0.0001), reading (p<0.0001), hearing function (p<0.0001) and socializing activities (p<0.02) show a strong correlation with cognitive function (whole model Rsq. 0.369, F ratio 50.1, p<0.0001). Using DCI for considering comorbidity the multivariate analysis model show similar results.

Conclusions: Physical activity, socialization and reading are associated to a better cognitive performance. It can be hypothesized that these factors can concur to cognitive decline prevention in elderly people.

Comunicazione presentata dagli Autori al 3° Congresso Nazionale FIMeG (Roma 23-25 giugno 2008 ). Atti Congressuali pubblicati dalla Casa Editrice Scientifica Internazionale (C.E.S.I. 2008,),via Cremona, 19 -
00161 Roma