Dynamic associations of cholinesterase inhibitors and memantine with cognitive trajectories in individuals with Alzheimer’s or mixed dementia: a real-world analysis using the quality registry SveDem

Alzheimer’s disease (AD) and mixed dementia (MxD) represent major public health concerns, yet there is limited real-world evidence on the long-term associations of commonly prescribed pharmacological treatments, particularly cholinesterase inhibitors (ChEIs) and their combination with memantine. This study aims to evaluate the long-term, time-varying associations of ChEIs and memantine on cognitive decline in a large, nationwide cohort of individuals diagnosed with AD or MxD. This observational study utilized data from the Swedish registry for cognitive/dementia disorders (SveDem), analyzing 32,282 individuals diagnosed with AD or MxD between 2007 and 2022. Patients were followed for up to 11 years to track treatment patterns and cognitive trajectories. Initiation of ChEIs (donepezil, galantamine, or rivastigmine) or memantine within six months after an AD or MxD diagnosis and then the time-varying medications (ChEIs alone, memantine alone or memantine added on to ChEIs) within six months after each follow-up were analyzed. Linear mixed-effects model was used to assess cognitive decline measured by Mini-Mental State Examination (MMSE) score trajectories. Of the 32,282 participants (mean age 78.4 years; 61% women), 78.4% initiated treatment with ChEIs alone, and 21.6% with memantine alone. Over time, prescription patterns shifted from monotherapy to combination therapy, with donepezil and galantamine more likely to achieve 1 Defined Daily Dosages (DDD) than rivastigmine users. Memantine alone users experienced a yearly cognitive decline of 1.79 MMSE points (95% CI: −1.85, −1.73). Compared with memantine users, patients on ChEIs alone declined 0.65 points less per year (95% CI: −0.59, −0.72), while those on combination therapy declined 0.19 points less per year (95% CI: −0.10, −0.28). Among ChEIs, donepezil users experienced an annual decline of 1.02 points (95% CI: −1.06, −0.98). In comparison, galantamine users declined an additional 0.05 points per year (95% CI: −0.11, 0.01), and rivastigmine users declined an additional 0.17 points per year (95% CI: −0.24, −0.11), relative to donepezil. In this large cohort of patients with AD or MxD, ChEIs users alone, particularly donepezil and galantamine, showed slower cognitive decline compared to users of memantine or combination therapy users. While differences were modest, the results contribute to a better understanding of treatment trajectories in routine clinical practice.