Revised Estimate Increases Global H1N1
Mortality Figures by Factor of 15
By Amesh A. Adalja, MD, FACP, June 29, 2012
The previously reported number of deaths due to laboratory-confirmed H1N1 (18,500) during the 16 months of the 2009-2010 pandemic (April 2009-August 2010)1 is widely acknowledged to be a gross underestimate because most flu patients were not tested. Evidence of this lack of data is that less than 12% of laboratory-confirmed deaths were reported from Southeast Asia and Africa, where 38% of the global population resides.1 Since realistic planning for future pandemics relies on accurate statistics, better estimates of the actual disease burden of the H1N1 pandemic are needed. Dawood and colleagues, representing an international collaboration headed by the CDC, recently reported the findings of an improved modeling study that may provide a more accurate estimate of global deaths from the H1N1 pandemic.2
This improved model used the symptomatic attack rate from high, middle, and low income countries multiplied by the symptomatic case fatality ratio from high income countries. The data on symptomatic attack rates were obtained from 17 sites in 13 countries in which visits to outpatient facilities for influenza-like illness were tabulated. To enhance precision, the attack and fatality rate estimates were stratified by age, country income levels, and WHO estimates for mortality from lower respiratory tract infections. Cardiovascular mortality and years of life lost were also calculated. A Monte Carlo simulation was then performed.2
15-fold Increase in Estimated Deaths
Using the new model, respiratory related influenza deaths were estimated at 201,200—a number 15 times the original estimated number of laboratory-confirmed cases. The greatest percentage of deaths (29%) was estimated to have occurred in Africa, which had a mortality rate 2 to 4 times higher than other regions. Not surprisingly, the highest percentage of deaths (65%) occurred in people aged 18 to 64 years. It is known that influenza is also associated with increased cardiovascular mortality. When estimated excess cardiovascular mortality was added, the estimated number of deaths rose to 284,400, with 9.7 million years of life lost (a mortality rate of 0.001-0.011).2 The estimated number of years of life lost was approximately 4 times greater than the estimate for a normal influenza season.
Accurate Estimates Needed
Despite the common perception that 2009 H1N1 was a mild pandemic, it exacted a substantial global toll in terms of mortality and years of life lost. The results of this study underscore the importance of accurate calculation of mortality, for these numbers will provide the basis for future influenza planning. A grossly underreported burden of disease could foster complacency in influenza planning, and a grossly overestimated number would also be problematic.2
Though the data presented by Dawood and colleagues is limited by the availability of accurate estimates of attack rates and fatality ratios, this model provides estimates that closely approach actual rates of mortality and have been arrived at through reliable methodology. Finally, this study demonstrates that inadequate data from specific geographic locales where high burdens of illness occur dramatically affect mortality rates.
Viboud C, Simonsen L. Global mortality of 2009 pandemic influenza A H1N1. Lancet Infectious Diseases 2012. doi:10.1016/S1473-3099(12)70152-4.
Dawood FS, Iuliano AD, Reed C, et al. Estimated global mortality associated with the first 12 months of 2009 pandemic influenza A H1N1 virus circulation: a modelling study. Lancet Infectious Diseases 2012. doi:10.1016/S1473-3099(12)70121-4