Systematic review and economic decision modelling for the prevention and treatment of influenza A and B

摘要: Objectives To establish the clinical and cost-effectiveness of amantadine, oseltamivir zanamivir compared to standard care for treatment prevention influenza. Data sources Electronic databases. Reference lists identified articles key publications. Relevant trials. Review methods A systematic review meta-analysis randomised evidence was undertaken investigate effectiveness prophylaxis use influenza B. An additional amantadine in children elderly also undertaken. Economic decision models were constructed examine cost-utility alternative strategies treating preventing and/or This informed by reviews outlined above information where required. Results The found that reduced median duration symptoms positive group 1.38 days otherwise healthy adult population, 0.5 day high-risk 1.5 population. 1.26 days, 1.99 1.3 similar groups inhaled zanamivir. relative risk reduction between approximately 75 90% 70 depending on strategy adopted population under consideration. For economic model a base case focussed primarily health benefits generated shortening period illness. that, care, estimated cost per quality-adjusted life year ranged from pound 6190 31,529 adults, 4535 22,502 'high-risk' group, 6117 30,825 children, 5057 21,781 residential included valuations effects pneumonia (and otitis media children's model) based observed rates However it does not include hospitalisations as only very limited data available antivirals hospitalisation rates. As mortality rates, deaths rare trials neuraminidase inhibitors (NIs). Therefore, suitable these sources. avoided costs are potentially important we carried out sensitivity analysis involved extrapolating reductions pneumonias NI deaths. In all four NIs is substantially improved this extrapolation. prophylaxis, antiviral drugs with vaccination preventative strategies. cases ratios either low or cost-saving. relatively unfavourable, there scenarios relating an could be cost-effective. Conclusions varies markedly intervention target populations. estimate sensitive variations certain parameters model, example proportion influenza-like illnesses literature used inform spans many decades hence great caution should exercised when interpreting results indirect comparisons model. Further making direct would valuable verify model's findings.