Revisiting influenza deaths estimates--learning from the H1N1 pandemic.

A major contributing factor to some of the criticisms of the management of the H1N1 pandemic was that its mortality impact was considerably less than had been predicted. Pre-pandemic global estimates of death were often in the millions while the ultimate death toll has been reported in the thousands. The large discrepancy in the predicted and reported death tolls suggests that there is a potential problem in how influenza deaths are estimated, counted and compared. The impetus for developing a coordinated and robust response to an influenza pandemic was largely driven by estimates of its potential death toll. The WHO predicted that the next influenza pandemic could result in 2–7 million deaths in a ‘best case’ scenario.1 Other analyses suggested even higher death rates. An analysis based on vital registry data from 1918 to 1919 predicted that if the next pandemic behaved like the 1918 pandemic, there would be ∼62 million deaths.2 Another estimate by a leading expert suggested that there could be 180–360 million deaths.3 According to the Canadian influenza pandemic plan, the lower bound of the 95% confidence interval for estimated Canadian deaths from a pandemic with a 35% attack rate was 24 603 (Table 1).4 View this table: Table 1 Sample of some quoted estimates of death from next influenza pandemic, seasonal influenza death estimates and reported deaths from H1N1 Yet, at the completion of the pandemic, the worldwide death toll was 18 156, the Canadian death toll 428.5,6 In other words, the most conservative estimates of Canadian deaths in one scenario was approximately 6000 greater than the worldwide death toll. What is …