Cost-effectiveness of escitalopram for generalized anxiety disorder in Canada.

BACKGROUND Generalized Anxiety Disorder (GAD) is a common chronic disease with a lifetime prevalence estimated to range from 4.2% to 12.7%. GAD places a substantial burden upon patients and healthcare resources. OBJECTIVE To determine the cost-effectiveness of escitalopram for GAD in a Canadian primary care setting from two perspectives [Ministry of Health (MoH) and society (SOC)]. METHODS A 24-week decision-analytic model was constructed using Data/TreeAge software. Patients were treated with escitalopram or generic paroxetine. Clinical rates were determined from the literature; expert opinion guided model pathway development. Effectiveness was measured as 'symptom-free days' (SFDs). Analyses from MoH perspective focused on direct costs of treatment (drugs, physician visits), while SOC also accounted for indirect costs associated with workdays lost due to GAD. Unit costs of healthcare services and wage rates were obtained from standard Canadian sources (2005 Canadian $ values). Cost-effectiveness was expressed as the incremental cost-effectiveness ratio (ICER). Extensive one-way and probabilistic sensitivity analyses were conducted. RESULTS Escitalopram was associated with higher expected number of SFDs than paroxetine (86.4 vs. 77.0 SFD, respectively). From the MoH perspective, expected costs were Can$724 and Can$663 for escitalopram and paroxetine arms, respectively, resulting in the ICER for escitalopram vs. paroxetine of Can$6.56/SFD (Can$2362/symptom free year). From the SOC perspective, escitalopram dominated paroxetine as more effective on SFDs and less costly. Sensitivity analyses demonstrated robustness of the model. Limitations include the absence of comorbidities, which are common in practice, lack of long-term data, and assuming that dropouts in trials reflect those in practice. CONCLUSION Escitalopram was found to be cost-effective compared with paroxetine in treatment of GAD from the Canadian MoH perspective, and dominating paroxetine from the SOC perspective. Therefore, a possible advantage may exist at the population level in the treatment of GAD with escitalopram in Canada.