Could a federal program to promote influenza vaccination among elders be cost-effective?

BACKGROUND Influenza-related mortality predominately and disproportionately impacts the elderly. Rates of annual influenza vaccination among the elderly are approximately 65%, far below the Healthy People 2010 target of 90%. We estimated the cost-effectiveness of a 10-year federal program to promote influenza vaccine, intended to increase vaccination rates among persons > or = 65 years old. METHODS Published estimates regarding influenza-associated mortality rates and vaccine efficacy among the US elderly were used to calculate the number needed to vaccinate (NNV) to prevent one all-cause death due to influenza, as well as the mortality reduction expected from increased vaccination rates. The costs per life-year saved were estimated for a hypothetical federal promotional campaign, patterned after a direct-to-consumer (DTC) advertising program (2006-2015). The base case scenario presumed a 25-percentage-point increase in vaccination rates to 90%; in sensitivity analyses, we examined programs that increased rates by 10-20 points. RESULTS The base case NNV was 1116 (95% CI: 993-1348). Over the 10-year DTC-style influenza vaccine promotion program, 6516 (5576-7435) elderly lives would be saved. The incremental cost-effectiveness (C/E) of the program was dollar 16,300 (dollar 11,347-dollar 25,174) per life-year saved in 2006 and increased to dollar 199,906 (dollar 138,613-dollar 307,423) per life-year saved by 2015. Overall, the C/E for the 10-year program was dollar 37,621 (dollar 32,644-dollar 43,939) per life-year saved. Programs that yielded a 15-percentage-point increase or less in vaccination rates would have C/E values exceeding dollar 50,000 per life-year saved and save fewer than 4000 total lives. CONCLUSIONS DTC-style promotional campaigns for influenza vaccine among elders may represent a cost-effective strategy for the federal government to pursue as a means of increasing elders' vaccination rates and reducing influenza-related mortality.