A population-based study of dosing and persistence with anti-dementia medications

Cholinesterase inhibitors and memantine are the mainstay of pharmacological intervention for the cognitive symptoms of Alzheimer's disease (AD). This study assessed the adequacy of dosing and persistence with AD medications and the predictors of these variables in the 'real world' (outside the clinical trial setting).

The Health Service Executive-Primary Care Reimbursement Services prescription claims database in the Republic of Ireland contains prescription information for 1.6 million people. Patients aged >70 years who received at least two prescriptions for donepezil, rivastigmine, galantamine and memantine between January 2006 and December 2010 were included in the study. Rates of dose-maximisation were recorded by examining the initiation dose of each AD drug commenced during the study period and any subsequent dose titrations. Non-persistence was defined by a gap in prescribing of more than 63 consecutive days. Predictors of dose-maximisation and non-persistence were also analysed.

Between January 2006 and December 2010, 20,729 patients aged >70 years received a prescription for an AD medication. Despite most patients on donepezil and memantine receiving a prescription for the maximum drug dose, this dose was maintained for 2 consecutive months in only two-thirds of patients. Patients were significantly more likely to have their doses of donepezil and memantine maximised if prescribed in more recent years (2010 vs. 2007). Rates of non-persistence were 30.1 % at 6 months and 43.8 % at 12 months. Older age [75+ vs. <75 years; hazards ratio (HR) 1.16, 95 % confidence interval (CI) 1.06-1.27] and drug type (rivastigmine vs. donepezil; HR 1.15, 95 % CI 1.03-1.27) increased the risk of non-persistence. Non-persistence was lower for those commencing therapy in more recent years (2010 vs. 2007; HR 0.81, 95 % CI 0.73-0.89, p < 0.001) and for those on multiple anti-dementia medications (HR 0.59, 95 % CI 0.54-0.65, p < 0.001). Persistence was significantly higher when memantine was co-prescribed with donepezil (p < 0.0001).

Future studies should explore the reasons underlying non-persistence and failure to maintain dose-maximisation in patients on AD medications. There may be scope to improve the dosing and persistence with these medications in the community.