Nutrition and cognitive deficit in the elderly: a population study

Objective: To investigate the association between a healthy diet indicator and the prevalence of cognitive impairment in the elderly. Design: Cross-sectional study. Setting: Population based. Subjects: A total of 1651 subjects (560 men and 1091 women) including every body aged 70 y or more, and a random sample of people (about 40%) aged 65 - 69 y resident in four rural towns in the province of Pavia, Italy in 1992- 1993. Interventions: The healthy diet indicator based on the WHO guidelines for t he prevention of chronic diseases was calculated as reported by Huijbregts et al (1998; Eur. J. Clin. Nutr. 52, 826-831). Food intake was estimated by means of a 180-item food-frequency questionnaire and nutrient intake was c alculated using the food composition database compiled for epidemiologic st udies in Italy. The cognitive function was categorized into four levels nor mal cognition, mild, moderate and severe cognitive deficit-according to the neuropsychological test score. The relationship beween the dietary and the ordinal cognitive function variables was studied using the proportional-od ds model. Results: After adjustment for age, sex, education, total energy intake, cig arrette smoking, alcohol consumption and physical activity, a better health y diet score was associated with a lower prevalence of cognitive deficit. T he cumulative odds ratio was 0.85 (95% Cl 0.77-0.93). Conclusions: Our results suggest an association beween a globally satisfact ory diet and better cognitive performance in the elderly. However, the spec ific aspects of a 'healthy diet' for the elderly should be clarified. Sponsorship: National Research Council (Italy), 'Invecehiamento' Project no . 95.01048.PF40.