Donepezil is the first and only acetylcholin- esterase inhibitor (AChEI) that is licensed for the treatment of severe Alzheimer disease (AD) 1 , but the maintenance of continuous treatment with AChEIs in advanced stages of the disease is being questioned. In the recently published DOMINO-AD study 2 , Howard et al. investigated how discontinuation of donepezil treatment for patients with moderate to severe AD affects their nursing home placement (NHP). The results provide some evidence that withdrawal of the drug increases the risk of NHP, but the analysis raises questions about the management of these patients. AD is the most common cause of demen- tia in the ageing population. The progressive course of the disease results in cognitive and functional decline that has several effects: it influences quality of life for patients and caregivers, successively increases the need for informal care and, in advanced stages of the disease, necessitates costly institutionalization that drains social care resources. Data from the large Swedish Dementia Registry (SveDem) show that >80% of patients with AD are treated with anti-dementia drugs 4 . Symptomatic treatment with AChEIs and/or the N-methyl-d-aspartate receptor antagonist memantine has been shown to improve or stabilize scores on cognitive and functional measures, and on measures of global clin- ical change, regardless of disease severity 3 . Donepezil is a potent, selective, noncompet- itive and rapidly reversible AChEI 1 . The effi- cacy of donepezil alone or in combination shown that treatment with donepezil is either cost-neutral or cost-saving 6 , but in an inde- pendent societal study in the UK, the AD2000 Collaborative Group reported that improve- ments in functional ability after treatment with donepezil would not delay NHP sufficiently to justify the costs of the drug 7 . The aim of the DOMINO-AD study was to determine how the NHP of patients who were receiving donepezil for moderate to severe AD was affected by continuation or discontinuation of donepezil treatment and either starting or not starting memantine treatment 2 . The study was a 1-year rand- omized double-blind placebo-controlled multicentre trial with an additional 3-year double-blind follow-up. The trial involved 15 secondary care memory centres in England and Scotland. In contrast to previous clinical trials of AChEIs, the inclusion and exclusion criteria for the DOMINO-AD study were relatively unselective to ensure that results were generalizable. Discontinuation of donepezil treatment was associated with an increased risk of NHP within 12 months; the risk was highest dur- ing the first 6 months. Beyond 12 months, patients who discontinued donepezil treat- ment were at no higher risk of NHP than were patients who continued treatment. Memantine, either alone or in combination with donepezil, had no effect on the risk of NHP. The delay of NHP among patients who continued donepezil treatment seems to have been driven by better cognition and ability to perform activities of daily living than in patients who discontinued donepe- zil treatment: the mean difference in scores between these groups were 1.9 points on the Standardized Mini-Mental State Examination, and 3 points on the Bristol Activities of Daily Living Scale. The results indicate that pro- longed treatment of six patients with severe dementia is needed to prevent one NHP. The study has been criticized for being under- powered for analysis of the secondary out- come of NHP 8 . Nevertheless, by reporting on the effects of AChEI withdrawal, the authors of DOMINO-AD have opened up a debate around whether the effects of AChEI therapy on NHP provide sufficient benefit to justify its continuation even in advanced stages of the disease, and whether combination treatment has any additional benefit. with memantine has been demonstrated by several short-term (≤24 months) randomized double-blind placebo-controlled clinical tri- als, and by long-term extension trials and observational studies 1 . Despite the demonstrated efficacy of AChEIs, whether or not treatment with them should be continued in the advanced stages of AD is still a matter of debate. In addition, there is only weak evidence for a clinical benefit of combination therapy with AChEI and meman- tine, the most consistently used approach for the management of dementia-related behavioural symptoms in advanced AD 5 . Methodological problems hinder the design- ing of long-term pragmatic trials that could address these issues, and prevent clinically meaningful outcomes and efficacy measures in severe AD from being defined. Furthermore, many regulatory authorities require not only evidence of a drug effect, but also of the cost effectiveness of a drug before they permit the development of clinical practice guidelines and reimbursement recommendations. NHP is an important milestone in the course of dementia, marks a shift from moder- ate to severe disease and dramatically increases the cost of care. Several modelling studies have ALZHEIMER DISEASE Donepezil and nursing home placement — benefits and costs Vesna Jelic and Bengt Winblad The recent DOMINO‑AD trial suggests that continued treatment with donepezil delays nursing home placement for patients with severe Alzheimer disease, but more work is needed to support strong conclusions about whether the benefits outweigh the costs. Refers to Howard, R. et al. Nursing home placement in the Donepezil and Memantine in Moderate to Severe Alzheimer’s Disease (DOMINO‑AD) trial: secondary and post‑hoc analysis. Lancet Neurol. http://dx.doi.org/10.1016/ S1474‑4422(15)00258‑6 Nursing home placement … is an important milestone in the course of dementia NATURE REVIEWS | NEUROLOGY VOLUME 12 | JANUARY 2016 NEWS & VIEWS © 2016 Macmillan Publishers Limited. All rights reserved