A Markov Model of the Cost Effectiveness of Olanzapine Treatment for Agitation and Psychosis in Alzheimer’s Disease

Antipsychotics have long been used to treat agitation and psychosis related to Alzheimer’s disease, but in a limited fashion because of troubling adverse effects. The new atypical antipsychotics are thought to be at least as effective as first-generation drugs and to cause fewer adverse effects. These drugs, however, are currently not US FDA-approved for use among elderly demented subjects due to a slight increase in the risk of death and serious cardiovascular events within this population. However, their favourable adverse effect profile has led many physicians to prescribe these drugs as first-line therapy for behaviourally disturbed patients with Alzheimer’s disease. Clinical trials to evaluate the use of atypical antipsychotics have produced varying results, and clarity has not yet been achieved. Thus, a quantitative summary of the risks and benefits may help inform complex decisions that must be made in this area. In this study we set out to compare the expected costs and outcomes for a community-dwelling cohort of patients with Alzheimer’s disease with agitation and/or psychosis who are (a) untreated, or (b) treated with olanzapine.

We constructed a Markov state-transition model using the best published data from several sources for Alzheimer’s disease patient progression and treatment. This model allowed us to compare the expected costs and outcomes associated with olanzapine treatment compared with no treatment for a synthetic cohort of US adults aged ≥65 years. The model cycles every 6 months and continues until all patients die from Alzheimer’s disease progression or from co-morbid conditions. Outcome estimates included the incremental cost-effectiveness ratio (ICER) and cost per quality-adjusted life-year (QALY) gained. The robustness of the estimates was examined by sensitivity analyses of key parameters, including cost of care, olanzapine effectiveness and Alzheimer’s disease progression rates.

Results indicated that olanzapine was a cost-effective treatment for agitation and psychosis related to Alzheimer’s disease when compared with no treatment (ICER <$US50 000). In addition, sensitivity analyses demonstrated that olanzapine remained cost effective despite multiple variations of several parameters, both alone and concurrently.

Olanzapine treatment for agitation and psychosis related to Alzheimer’s disease is cost effective when compared with no treatment. Further analysis should be performed as atypical antipsychotics become generic, as more information on health utilities in the Alzheimer’s disease population becomes available, and to compare atypical antipsychotics with first-generation antipsychotics.