Lack of association between proton pump inhibitor use and decline in cognitive performance in the ELSA-Brasil cohort.

Abstract

The study aims to investigate the longitudinal association of use and time of use of proton pump inhibitors (PPIs) with decreased performance in three cognitive tests. Prospective cohort study included 7115 participants with mean age of 58.9years at baseline (2008-2010) who participated in the second wave (2012-2014) of ELSA-Brasil (average interval between visits = 3.9years (range: 1.7 to 6.0years)). Cognitive performance was assessed by tests of memory, phonemic and semantic verbal fluency, and the trail making test, applied to both waves. Associations with the use and time of use of PPIs at baseline were investigated using linear models with mixed effects after adjusting for confounding factors. At baseline, 7.4% (529) of the participants used PPIs on a regular basis. After all adjustments, the interaction term use of PPI × age was not statistically significant for the cognitive tests evaluated, indicating that the use of PPI at baseline was not associated with a more accelerated decline in cognitive performance between waves. The interaction term PPI use × age was not statistically significant, in any of the categories of medication use time, any of the cognitive function tests evaluated, indicating that PPI use time is not associated with decrease in cognitive scores as the time interval between visits increases. In this cohort middle-aged and elderly adults, after average interval of 3.9years (relatively short time to detect cognitive decline in a young cohort), the use and time of use of PPIs at the beginning of the study were not associated with a decline in cognitive performance in these tests between visits.