Objective: The aim of this study was to develop a generic treatment algorit
hm for influenza and influenza-like illness (ILI) that could be used to est
imate the costs and outcomes of current and new treatments for influenza in
different countries for different patient subgroups.
Methods: A series of possible treatment pathways was identified and the pro
babilities of different patient subgroups following each pathway were estim
ated by using the published literature. The health outcomes and health serv
ice use and unit costs for each pathway were estimated from trial data and
standard data sources. An interactive computer model was created, the base-
case input parameter values were assigned, and estimates of the current cos
ts of influenza and ILI in different population subgroups estimated. Sensit
ivity analyses were performed by changing input parameter values.
Results: The average healthcare cost of influenza and ILI per person in the
US was $US72 for the general population and $US330 for a high risk populat
ion (1997 values). The average total cost per patient (healthcare cost plus
productivity losses) was $US320 for the general population and $US546 for
a high risk population. These costs are sensitive to changes in the proport
ion of patients visiting a physician and to the proportion of patients hosp
italised with complications of the disease. Days to alleviate major symptom
s and other health outcome measures are sensitive to the percentage of pati
ents who receive antiviral therapy as well as to the efficacy of this thera
py.
Conclusions: The costs and health outcomes of influenza and ILI depend on t
he extent to which patients visit a physician, the use of antiviral drugs,
and the incidence of complications requiring hospital care. The computer mo
del will allow decision-makers to assess the cost effectiveness and the pot
ential budget impact of new antivirals for treating influenza.