Background: The impact of influenza on mortality in countries in subtropical and tropical regions is poorly quantified. Estimation of influenza-related illness in warm-climate regions is more difficult, because the seasonality of virus circulation is less well-defined. Partly as a result of these factors, influenza vaccine is grossly underutilized in the tropics, even for individuals >or=65 years of age.
Methods: Weekly numbers of deaths were modeled by Poisson regression, and excess deaths attributable to influenza in Hong Kong were estimated for 1996-1999. Comparison of weekly mortality during periods of influenza predominance and periods of low influenza activity was used to derive an alternative estimate of influenza-associated mortality.
Results: Estimates derived from the Poisson model indicated that influenza resulted in 7.3 deaths per 100,000 population per year (95% confidence interval [CI], 3.1-11.4) from cardiorespiratory disease among individuals aged 40-65 years and 102.0 deaths per 100,000 per population per year (95% CI, 61.2-142.7) among individuals aged >or=65 years. Although respiratory diseases accounted for the majority of influenza-related deaths, influenza also contributed to 13.8% (95% CI, 4.8%- 22.7%) and 5.3% (95% CI, 1.2%-9.3%) of deaths related to ischemic heart disease.
Conclusion: Influenza is associated with deaths due to ischemic heart disease as well from respiratory diseases. Overall influenza-associated mortality in a region with a warm climate, such as Hong Kong, is comparable with that documented in temperate regions. The need for influenza vaccination in tropical regions needs to be reassessed.