Antidementia drugs and nursing home placement: a systematic review and meta-analysis

The association between antidementia drugs (ADDs) use and the risk of nursing home placement (NHP) remains inconclusive. This study aimed to investigate the effects of ADDs, including cholinesterase inhibitors (CEIs) and memantine, on NHP among individuals with dementia.

A systematic search of PubMed, Embase, Cochrane Library and ClinicalTrials.gov was conducted up to 16 March 2024 and updated on 25 August 2025. Randomized controlled trials (RCTs) or observational studies investigating the use of ADDs and NHP were included. Hazard ratios (HRs) with 95% confidence intervals (CI) were pooled using the DerSimonian-Laird random-effects model.

Of 1,373 records, three RCTs and nine observational studies were included, encompassing different comparators. Meta-analyses were conducted separately by study design. In RCT, the single trial comparing any ADDs versus non-use (N = 1; donepezil vs placebo) showed no significant difference in NHP. In head-to-head RCT comparisons, one trial suggested a non-significant trend toward higher NHP risk with memantine versus donepezil. Among observational studies, a meta-analysis of five observational studies suggested a lower NHP risk with any ADDs in Alzheimer's disease (pooled HR = 0.43, 95% CI: 0.32-0.58, I² = 40.39%, p < 0.001) and mixed dementia (pooled HR = 0.84, 95% CI: 0.72-0.97, I² = 0.00%, p = 0.019). Pooled observational head-to-head comparisons likewise showed no significant differences between donepezil and rivastigmine (n = 3), donepezil and galantamine (n = 2), or CEIs + memantine versus CEIs monotherapy (n = 2). These observational estimates may be affected by residual confounding and other biases and should be interpreted with caution. By GRADE, certainty was low for the RCTs evidence and very low for the pooled observational studies of ADDs or CEIs versus non-use, and for head-to-head and combination therapy comparisons.

RCTs, rated as low-certainty evidence, suggest that ADDs may have no effect on NHP, whereas observational studies, rated as very low-certainty evidence, suggest that ADDs may be associated with a reduced risk of NHP among individuals with dementia. Given the overall uncertainty, high-quality, adequately powered prospective trials with longer follow-up are required to clarify these associations and to assess whether a causal relationship exists.