Dietary glycemic load and risk of cognitive impairment in women: findings from the EPIC-Naples cohort

Cognitive impairment is a common cause of morbidity in the elderly. The relationship between dietary habits and cognitive impairment in a female population living in the metropolitan area of Naples, in the Mediterranean part of Italy, has been evaluated in the Naples EPIC prospective cohort study. The study cohort, enrolled between 1993 and 1997, is composed of 5062 women aged 30-69 years. At time of enrolment anthropometric measures were performed and information about socio-demographic details, clinical data, lifestyle and dietary habits were collected. During 2008 and 2009, women 65 years of age or older received a telephone interview to evaluate cognitive status (TICS); the derived score was used as proxy of cognitive impairment. Analyses were carried out on 1514 participants. Linear regression model showed negative association between TICS score and, respectively, age at baseline (beta = -.31, 95 % CI -.34, -.24), body mass index (BMI) (beta = -.08, 95 % CI -.16, -.01), and glycemic load (GL) (beta = -.02, 95 % CI -.03, -.01), whereas education level (beta = 0.62, 95 % CI .56, .69) showed positive association. A logistic regression model, used to evaluate determinants of the low cognitive score (TICS score a parts per thousand currency sign 15, 1st tertile), confirmed association for previous variables [age (OR 1.1, 95 % CI 1.08, 1.15); BMI (OR 1.03, 95 % CI 1.001, 1.07); GL (OR 1.005, 95 % CI 1.001, 1.011); education level (OR .82, 95 % CI .79, .84)] with, in addition, type II diabetes (OR 1.85, 95 % CI 1.014, 3.4). This study indicates that GL may play a role in determining risk of cognitive impairment, besides age, BMI, education and diabetes.