The aim of this study was to determine the cost effectiveness of influenza vaccination for healthy people aged 65-74 years living in the UK. People without risk factors for influenza (chronic heart, lung or renal disease, diabetic, immunosuppressed or those living in an institution) were identified from 20 general practitioner (GP) practices in Liverpool in September 1999. 729/5875 (12.4%) eligible individuals were recruited and randomised to receive either influenza vaccine or placebo (ratio 3:1), with all participants receiving 23-valent-pneumococcal polysaccharide vaccine unless already administered. The primary analysis was the frequency of influenza as recorded by a GP diagnosis of pneumonia or influenza like illness. In 2000, the UK vaccination policy was changed with influenza vaccine becoming available for all people aged 65 years and over irrespective of risk. As a consequence of this policy change, the study had to be fundamentally restructured and only results obtained over a one rather than the originally planned two-year randomised controlled trial framework were used. Results from 1999/2000 demonstrated no significant difference between groups for the primary outcome (relative risk 0.8, 95% CI 0.16-4.1). In addition, there were no deaths or hospitalisations for influenza associated respiratory illness in either group. The subsequent analysis, using both national and local sources of evidence, estimated the following cost effectiveness indicators: (1) incremental NHS cost per GP consultation avoided = 2000 pound sterling; (2) incremental NHS cost per hospital admission avoided = 61,000 pound sterling; (3) incremental NHS cost per death avoided = 1,900,000 pound sterling and (4) incremental NHS cost per QALY gained = 304,000 pound sterling. The analysis suggested that influenza vaccination in this population would not be cost effective.