Potentially inappropriate medications and cognitive performance: cross-sectional results from the ELSA-Brasil study.

OBJECTIVES: Potentially inappropriate medications (PIM), especially those with potential effects on the central nervous system, can increase the risk of cognitive impairment. We investigated the association of the use of PIM and PIM that may impair cognition (PIM-Cog) with cognitive performance among older adults. METHODS: In this cross-sectional study with 2,626 participants, PIM and PIM-Cog were defined by the 2019 American Geriatrics Society Beers criteria. We calculated global cognition and memory, verbal fluency, and Trail Making Test B version (TMT-B) z-scores. Linear regression models adjusted for sociodemographic and clinical variables were used to investigate the association between PIM and cognition. RESULTS: 27% and 7% of the sample (mean age = 65.1 ± 4.1 years old, 54% women, and 61% White) used at least one PIM and PIM-cog, respectively. PIM was associated with poor performance in the TMT-B (ß = -0.17, 95% Cl = -0.29; -0.05, p = 0.007). PIM-Cog was also associated with poor TMT-B performance (ß = -0.08, 95% Cl = -0.15; -0.01, p = 0.025). CONCLUSION: The use of PIM and PIM-Cog was associated with poor executive function among older adults. The review of PIM use and the deprescription of these drugs may be an effective way to improve cognitive function.

Anticholinergic burden and cognitive performance: cross-sectional results from the ELSA-Brasil study.

OBJECTIVES: Using multiple drugs with anticholinergic properties is common and might lead to cumulative anticholinergic toxicity and increased risk of cognitive impairment. Therefore, we sought to investigate the association between the Anticholinergic Cognitive Burden (ACB) Scale and cognitive performance among middle-aged and older adults. METHODS: In this cross-sectional study with 13,065 participants from the baseline visit of the Brazilian Longitudinal Study of Adult Health (ELSA-Brasil), mean age was 51.7 ± 9.0 years old, 55% women, and 53% white. The ACB was calculated based on the medications in use. We investigated the association of ACB with global cognition and memory, verbal fluency (VF), and trail-making test version B (TMT-B) z-scores, using multiple linear regression models adjusted for sociodemographic and clinical variables. RESULTS: Overall, 16% of participants had an ACB score greater than 0. ACB was associated with poor cognitive performance in all tests in crude analysis. After adjustment for sociodemographic characteristics, the association remained significant for the global cognitive score, as well as the memory and the TMT-B z-scores. However, after further adjustments for clinical variables, only trend associations of ACB with poor memory (ß = - 0.02, 95% Cl = - 0.05, 0.00, p = 0.056) and the TMT-B z-scores (ß = - 0.02, 95% Cl = - 0.04, 0.00, p = 0.054) were found. In stratified analyses by age groups, ACB was associated with poor cognitive performance on the TMT-B (ß = - 0.03, 95% Cl = - 0.05, - 0.01, p = 0.005) in individuals aged less than 65 years old. CONCLUSION: Although the ACB was associated with poor executive function only among middle-aged adults in adjusted analysis, residual confounding may partly explain our results.