Background: Observational studies have suggested delay in nursing home placement (NHP) with dementia drug treatment, but an earlier randomised trial in patients with mild to moderate Alzheimer’s disease (AD) showed no effect. We investigated the effects of continuing or discontinuing donepezil and starting memantine on subsequent NHP in moderate to severe AD.
Methods: In the DOMINO trial (ISRCTN49545034) 295 community living patients with moderate to severe AD recruited from 15 centres in England and Scotland from February 2008 to March 2010 were randomised with double-blind placebo-control to continue donepezil (73), discontinue donepezil (73), discontinue donepezil and start memantine (76), or continue donepezil and start memantine (73) for 52 weeks. After 52 weeks choice of treatment was left to participants and their physicians. Place of residence was recorded at outcomes assessment points during the first 52 weeks of the trial and subsequently every 26 weeks for a further 3 years. Nursing home placement was an irreversible move from independent accommodation to a residential caring facility and was a secondary trial endpoint. Analyses restricted to the risk of placement in the first year of follow-up were post-hoc.
Findings: 162 patients (55%) underwent NHP within 4 years of randomisation. Numbers of NHPs were similar for all arms (36 in patients who continued donepezil, 42 who discontinued donepezil, 41 who discontinued donepezil and started memantine, and 43 who continued donepezil and started memantine). There was significant (p=0.010) heterogeneity of treatment effect over time with significantly more NHPs in the donepezil discontinuation group during the first year (HR 2.09 (95% CI, 1.29 to 3.39)) and no difference later (HR 0.89 (95% CI, 0.58 to 1.35)). Subsequent analyses focussed on the first year of the trial and on donepezil only were post-hoc. 1-year NHP risk was 17% higher (95% CI 6% to 28%) in patients allocated to discontinue donepezil compared to continuing donepezil. There was no effect of starting memantine compared to no memantine during the first year (HR 0.92 (95% CI 0.58 to 1.45)) or later (HR 1.23 (95% CI 0.81 to 1.87)); difference in 1-year NHP risk 1% (95% CI -12% to 10%).
Interpretation: Withdrawing donepezil in patients with moderate to severe AD increased the risk of NHP during 12 months of trial treatment, but made no difference to NHP over 4 years of follow-up. Decisions to stop or continue drug treatment at this stage should be informed by potential risks of withdrawal, even if the perceived benefits of continued treatment are not clear.