Cost-effectiveness of newer treatment strategies for influenza

Abstract

PurposeRecent advances in the diagnosis and treatment of influenza, such as rapid testing and neuraminidase inhibitor therapy, are available, but their place in clinical practice and their cost-effectiveness have not been determined. Materials and methodsTo estimate the cost-effectiveness of these newer interventions, we used a decision model that compared several influenza management strategies: no testing or treatment, amantadine or rimantadine treatment without testing, testing then amantadine or rimantadine treatment, neuraminidase inhibitor treatment without testing, or testing then neuraminidase inhibitor treatment. Antiviral therapy began within 48 hours in febrile patients with characteristic symptoms of influenza. We assumed that antiviral treatment did not change rates of influenza complication or mortality, and chose parameter values in the baseline analysis to bias slightly against antiviral treatment and toward testing strategies. ResultsIn the baseline analysis, testing strategies are more expensive and less effective than treatment strategies. Amantadine costs $9.06 per illness day avoided or $11.60 per quality-adjusted day gained. Compared with amantadine, zanamivir costs $198 per illness day avoided or $185 per quality-adjusted day gained, whereas oseltamivir costs $252 per illness day avoided or $235 per quality-adjusted day gained. In elderly patients who require reduced dosage, rimantadine costs $128 per quality-adjusted day gained compared with amantadine. In younger patients, amantadine is favored if the likelihood of influenza A is >67%; otherwise, neuraminidase inhibitors are favored. Testing strategies are more costly and less effective when the influenza probability is >30%. No testing or treatment is favored if the influenza probability is <32% and the influenza utility is >0.77. In elderly patients, amantadine is favored over rimantadine if the utility of medication side effects is >0.94. ConclusionAntiviral treatment of influenza without rapid testing is reasonable economically in febrile patients with typical symptoms during influenza season. The choice of antiviral agent depends on age, the likelihood of influenza A, and the willingness to pay per quality-adjusted day gained.