Risk of antipsychotic drug use in patients with Alzheimer's disease treated with rivastigmine

Background and Objective: Cholinesterase inhibitors may offer some improvement in the behavioural symptoms of Alzheimer's disease. The dual inhibitory mechanism of action of rivastigmine (inhibition of acetylcholinesterase and butyrylcholinesterase) may improve behavioural symptoms and may delay the need for antipsychotics. This study was conducted to investigate the effect of rivastigmine on the time to first antipsychotic drug use among patients with Alzheimer's disease, compared with patients with Alzheimer's disease not treated with a cholinesterase inhibitor. Design and Methods: This study used MarketScan((R)) research databases from 1 January 1999 to 31 March 2002. Patients were included if they: (a) were diagnosed with Alzheimer's disease on two occasions or filled a prescription for rivastigmine for the first time during the index period from 1 July 2000 to 31 December 2001; (b) were 65 years of age and older; (c) had continuous health and prescription insurance coverage during the entire study period; and (d) had not used an antipsychotic medication within 18 months prior to their index Alzheimer's disease prescription or diagnosis. The 'no cholinesterase inhibitor' group included patients who were newly diagnosed with Alzheimer's disease, but did not use any cholinesterase inhibitors. Chi-square, Student's t-, and log-rank tests were used to test differences in study variables between groups. Cox proportional hazards models were used to estimate predicted risk of first antipsychotic drug use. Results: The study included 497 patients in the rivastigmine group and 749 patients in the `no cholinesterase inhibitor' group. The rivastigmine group patients were younger compared with the `no cholinesterase inhibitor' group patients (p < 0.01). The overall usage of antipsychotics was considerably lower for patients taking rivastigmine (9.8%) compared with those not taking cholinesterase inhibitors (25.6%). Patients taking rivastigmine were 64% less likely (relative risk = 0.36; p < 0.0001) to take antipsychotics compared with patients not taking cholinesterase inhibitors, after adjusting for demographic covariates, comorbid conditions, and use of other CNS drugs and anticonvulsants. Age was the only other factor that influenced antipsychotic use; older patients were significantly more likely to start antipsychotics than younger patients. Conclusion: This study provides initial evidence that patients with Alzheimer's disease treated with rivastigmine have a reduced risk of initiating therapy with an antipsychotic drug compared with patients who receive no cholinesterase inhibitor treatment. These findings may imply that rivastigmine use could delay the onset of behavioural symptoms that require treatment with antipsychotic medications.