The cognitive neutrality of common antidiabetic medications in older adults: A propensity score-matched cohort analysis from the National Alzheimer's Coordinating Center database

Abstract

Background

Type 2 diabetes mellitus (T2DM) affects about 25% of adults aged 65 years and older in the United States. It is linked to faster cognitive decline and nearly double the risk of dementia. However, the long-term cognitive effects of different diabetes medications are not well understood, as observational studies are often impacted by various biases.

Objective

This study aims to compare cognitive changes over time among older adults with T2DM who are taking different diabetes medications.

Methods

We examined data from 54,631 participants (204,031 observations) in the National Alzheimer's Coordinating Center (NACC) database from 2005 to 2019. We applied propensity score matching to reduce bias and create balanced groups for medication comparisons. We used linear mixed-effects models to assess rates of cognitive decline in 18 areas, including overall cognition, memory, executive function, language, and daily functioning. We also ran sensitivity, mediation, and moderation analyses, applying corrections for multiple testing using the domain-specific false discovery rate (FDR).

Results

Out of 86 comparisons between medication and cognitive outcomes, 96.5% (83 out of 86) found no meaningful differences in rates of cognitive decline. The only exception was a sharper decline in functional abilities (CDR judgment, community affairs, home/hobbies) for insulin-sulfonylurea combinations compared to insulin-metformin combinations (p < 0.05; Cohen's d = 0.52–0.59; NNT = 23–32). Metformin did not show any cognitive benefit compared to no medication, sulfonylurea, or insulin alone. The length of time a person has had diabetes turned out to be a stronger predictor of cognitive decline than any specific medication (p < 0.001). These results were consistent through additional analyses and corrections for multiple testing.

Conclusions

The choice of diabetes medication has little effect on cognitive changes in older adults with T2DM, as 96.5% of comparisons showed no significant differences. The duration of the disease, rather than the type of medication chosen, has a bigger impact on cognitive outcomes. This suggests that starting and maintaining good blood sugar control is more important than the specific cognitive effects of different medications.