Factors that predict nursing home placement in patients with Alzheimer’s disease include worsening dementia, behavior disturbances, and caregiver burden. However, most families prefer to keep patients with Alzheimer’s disease at home as long as possible. In addition, delaying nursing home placement may reduce overall health care costs. Geldmacher and colleagues studied the role of the cholinesterase inhibitor donepezil in delaying time to nursing home admission.

The study was an observational follow-up to several randomized, placebo-controlled trials investigating the therapeutic benefits of donepezil. At the completion of these trials, patients were invited to continue open-label treatment for up to 240 weeks. Data, including nursing home placement, were recorded, as were the date and reason for placement. Four groups of patients were defined, depending on duration of donepezil use and dosage. These groups included the following: (1) minimal-use patients who received less than the minimum dosage, (2) early moderate-use patients who received adequate dosages but terminated early or did not participate in the open-label extension, (3) delayed-start patients who received adequate dosages only after entering the extension trial, and (4) maximal-use patients who received adequate dosages during the placebo-controlled phase and the extension. The principal outcome was time from entry in a placebo-controlled trial to nursing home placement. Different constructs were analyzed, depending on the duration of successful participation in a clinical trial.

A total of 671 patients with mild to moderate dementia were included in the analyses. Estimated median times to first dementia-related nursing home placement ranged from 44.7 months in the minimal-use group and 66.1 months in the maximal-use group. A delay in first dementia-related nursing home placement of 21.4 months was associated with donepezil use of at least 5 mg daily for at least 36 to 48 weeks. A delay of 17.5 months to permanent nursing home placement was associated with donepezil use of the same dosage and duration. Relative risk ratios (RRs) for nursing home placement in the moderate-use group were not significantly different from those in the minimal-use group but were significantly different from this group in the delayed-start and maximal-use groups. RRs for nursing home placement declined with longer donepezil use across all four groups, a trend that was consistent for the three cohort time constructs analyzed.

According to the results of this study, cumulative sustained exposure to effective doses of donepezil delays time to nursing home placement when placement is for longer than two weeks. Washout periods during the placebo-controlled trials attenuated donepezil’s effects, suggesting that treatment of Alzheimer’s disease with donepezil should be maintained before important benefits will occur. The authors doubt that selection bias influenced these results, given the similarity of patient characteristics and a follow-up period of up to eight years. Donepezil appears to benefit both cognitive function and behavioral symptoms.