Cost effectiveness analysis of TV campaign to promote flu shot among elderly

Objective: The influenza and pneumonia account for 90% of deaths among the elderly (65+), and ranked 7^th among all causes of death for all age groups in 2003. Previous research implied the potential effectiveness of a television (TV) flu shot campaign to increase vaccination rate and ultimately reduce the serious flu disease burden among the elderly. The objective of this study is to determine the incremental cost-effectiveness of "a flu shot campaign using a national TV network" compared with "no flu shot campaign" targeting US Medicare elderly.

Methods: We developed probabilistic decision models to estimate the incremental cost-effectiveness ratio (ICER) of a national TV campaign on flu shot, accounting for the different effectiveness across ethnicity groups. Since English speaking Hispanic and Spanish speaking Hispanic showed different effectiveness from TV campaign on flu shot, we split the ethnicity groups into two levels: 3 race level (White, African American, and Hispanic) and 4 race level (White, African American, English speaking Hispanic, and Spanish speaking Hispanic). Then, the estimation was applied separately for each race level. The intervention is the national TV campaign on flu shot during 4 month (September 1 ~ December 31) via 3 national TV networks, compared with no TV campaign. The total cost included only TV campaign cost (excluding the vaccination cost). The main outcome measure is ICER indicating incremental dollar value (in 2012 dollar value) per an additional Medicare elderly vaccinated as a result of the TV campaign. Monte-Carlo simulations were performed in hypothetical cohorts of 2012 US Medicare elderly. Data sources for model parameters included our original analyses of Medicare Current Beneficiary Survey (1999~2001), the literature, and the related websites.

Results: The national TV campaign on flu shot appears more cost-effective (i.e., ICER = $22.74 per vaccinated excluding vaccination cost, 95% confidence interval (CI) = ($13.77, $46.25)) in 4 race level, compared to other relevant interventions such as a reminder system (e.g., ICER = $58~$412 per additional patient vaccinated including vaccination cost) and a school vaccination clinic for children (e.g., ICER = $32.16 per vaccinated child excluding vaccination cost) in the literature. In 4 race level, the examined campaign was most cost effective among non-Hispanic white population (ICER = $17.90, 95% CI= ($11.42, $32.93)) and least cost effective among Spanish speaking Hispanic population (ICER = -$5.83, 95% CI= (-$23.57, -$2.38)). The flu shot TV campaign was expected to increase the race/ethnic disparity of flu vaccination coverage by 0.2% between the white and the African American, 5% between the white and the Hispanic (in 3 race level) and 3.2% between the white and the Spanish speaking Hispanic (in 4 race level).

Conclusions: Overall, the national TV campaign on flu shot was preferred (more cost effective), compared with other relevant interventions. The results can justify public expenditures to implement a future vaccination campaign using a national TV network.