PREDICTING THE RISK OF COGNITIVE DECLINE IN OLD AGE: THE ANU-ADRI SCORE IN THE INCHIANTI-STUDY

Purpose and methods: Self-report screening risk scores have been proposed to predict the risk of dementia. Aim of the present study was to assess the ability of ANU-ADRI Score (AAS) to predict Cognitive Decline Trajectories (CDTs) in the InCHIANTI cohort, which is a longitudinal epidemiological study including old age community-dwelling subjects. We considered participants with Mini Mental State Examination (MMSE) >=24 at baseline surveywho completed, survival at 9- years follow-up. AAS (modified according to data availability) was computed at baseline with modifications accordingly to data availability. CDTs were estimated by Latent Class Growth Curve Models of MMSE score change and ROC curves were utilized to assess the ability of AAS to predictivity in future CDT allocations. Results: 463 participants (age 71.327±4.62, 246 F) were included. We identified 4 clusters with different CDT:, labelled as rapid decliners (RD), slow decliners (SD), low functioning stable (LFS) and high functioning stable (HFS). Among AAS variables, the four groups differed significantly for age (lowest in HFS and LFS), education (highest in HFS), depressive symptoms (Center for Epidemiological Studies Depression ScaleCES-D>16, (less frequent in HFS), high cholesterol (more frequent in RD) and fish consumption (less frequent in SD). Baseline AAS was able to predict future cognitive decline, with an AUC of 0.76 for RD or DS vs. LFS or HFS and an AUC of 0.77 for RD vs. all other groupsCDTs. Conclusion: The use of AAS showed a satisfactory ability, in predicting trajectories of future cognitive decline in an older subjects cohort. This might be valuable to easily identify high-risk subjects who might be targeted by preventive interventions.