Anticholinergic/sedative drug burden predicts worse memory acquisition in older racially/ethnically diverse patients with type 2 diabetes mellitus

Abstract

Objective: Anticholinergic/sedative drug use, measured by the Drug Burden Index (DBI), is linked to cognitive impairment in older adults. Yet, studies on the DBI's association with neuropsychological functioning are lacking, especially in underserved groups at increased risk of cognitive impairment. We examined cross-sectional relationships between total DBI (DBI_T ) and an age-adjusted analogue (Adj DBI_T ) with the Repeatable Battery for the Assessment of Neuropsychological Status (RBANS) in diverse adults with type 2 diabetes mellitus (T2DM). Based on results of a prior study, we anticipated higher DBIs would be associated with worse memory at older ages.

Methods: One hundred five adults with T2DM (age = 57 ± 9 years, 65% female, 62% Black, 27% Hispanic/Latino, Hb_A1c = 7.8 ± 1.8) participated. Although memory outcomes were normally distributed, DBI_T values were positively skewed. Spearman correlations assessed their bivariate relationships with RBANS. Adjusting for comorbidities, polypharmacy, Hb_A1c , and education, we tested the moderating effect of age on DBI-RBANS associations at mean ±1 standard deviations of age.

Results: One third of the participants endorsed current sedative/anticholinergic use. Mean DBI_T was 0.385, and mean Adj DBI_T was 0.393 (ranges = 0.00-4.22). Drug burden negatively correlated with RBANS Immediate Memory (DBI_T r_s = -0.237, P = .013; Adj DBI_T r_s = -0.239, P = .014) but no other indices. There was a significant DBI*Age interaction; the negative effect of drug burden on Immediate Memory was significant for ages greater than or equal to 55 years old.

Conclusions: Sedative/anticholinergic drug exposure was prevalent in these diverse T2DM patients. Adjusting for covariates, greater drug burden was associated with worse memory acquisition among older adults only. Prospective studies should examine these relationships over time and assess whether dementia biomarkers affect the interaction.

Keywords: drug burden index; memory; polypharmacy; type 2 diabetes.

FIGURE 1

A, Regression lines representing the relationship between Total Drug Burden Index scores and Immediate Memory, moderated by age. Interaction depicted at the mean age ±1 standard deviation (SD) of age (b = −0.683, P = .011). B, Regression lines representing the relationship between Age-Adjusted Total Drug Burden Index scores and Immediate Memory, moderated by age. Interaction depicted at the mean age ±1 SD of age (b = −0.712, P = .007)