Background: Oseltamivir is effective in the treatment of influenza. Utilisation in The Netherlands is limited, but increasing. Objective: To estimate the cost effectiveness of oseltamivir treatment (vs symptom relief only) for patients with influenza-like illness (ILI) who are at increased risk for serious complications of influenza. Methods: A cost-effectiveness analysis was used, building on a previously developed model (decision tree) that was applied for evaluating influenza vaccination and pandemic preparedness plans. Three patient subgroups were assessed (elderly patients [aged >= 65 years] without chronic disease, elderly patients with chronic disease, and chronically ill, non-elderly patients). Inputs for the model were taken from various sources including a meta-analysis. A societal perspective was adopted and costs were expressed in E per life-year gained (year 2003 values). Life-years lost were discounted at 4% in accordance with Dutch guidelines. Deterministic and probabilistic sensitivity analyses were employed to assess the robustness of the results. Results: For chronically ill patients with ILI, visits to the GP for oseltamivir treatment were cost saving. For non-chronically ill elderly patients, incremental cost-effectiveness was estimated at E1759 per life-year gained. Cost savings and favourable cost effectiveness were robust in a deterministic and stochastic sensitivity analysis. Conclusion: Our model-based analysis suggests that at-risk people presenting with ILI to a GP could be offered oseltamivir at favourable cost effectiveness or even cost savings in the Dutch setting compared with symptom relief with analgesics only.