OBJECTIVES: Little information exists on whether nationwide efforts to reduce antipsychotic use among nursing home (NH) residents with Alzheimer's disease and related dementias improved mortality and hospitalization outcomes for residents. Our objective was to examine the effect of NH decreases in antipsychotic use on outcomes for residents with Alzheimer's disease and related dementias. DESIGN: Observational nationwide study that emulated a series of cluster randomized trials. SETTING AND PARTICIPANTS: Long-stay NH residents with Alzheimer's disease and related dementias in US NHs. METHODS: The study used data from Medicare claims to emulate cluster randomized trials in which NHs were assigned to either decrease or maintain/increase antipsychotic use. Outcome ascertainment for the first trial began on April 1, 2012 (ie, following the announcement of the National Partnership to Improve Dementia Care in NHs). The last day of follow-up was December 31, 2017. Outcomes measured included 12-month all-cause mortality, all-cause hospitalization, and hospitalization for stroke, myocardial infarction, fracture, and psychiatric conditions. Use of other psychotropic medications was also evaluated. Inverse-probability-of-treatment-weighted pooled Poisson regression models estimated covariate-adjusted risk ratios (RRs). RESULTS: The adjusted risks of death (RR, 1.01
95% CLs, 1.00, 1.01), all-cause hospitalization (RR, 1.00
95% CLs, 1.00, 1.01), and hospitalization for specific causes were similar between resident-trials in NHs that decreased vs maintained/increased antipsychotic use. Use of antidepressants, anxiolytic/sedative-hypnotics, anticonvulsant/mood stabilizers, and antidementia medications was slightly higher among resident-trials in NHs that decreased antipsychotic use. CONCLUSIONS AND IMPLICATIONS: Decreases in NH antipsychotic use do not appear to improve resident outcomes. Intensive initiatives focused predominantly on achieving a decrease in antipsychotic use may not be effective at improving mortality and hospitalization outcomes for residents with Alzheimer's disease and related dementias. These findings suggest the need for better strategies that incorporate safe and effective nonpharmacological or pharmacological alternatives for managing neuropsychiatric symptoms of dementia.