winter of 1918-1919 (Taubenberger and Morens, 2006). Retrospectively, the earliest cases seen in the United States appeared in the U.S. military.
The strain of influenza responsible for the 1918 pandemic was particularly deadly, with a fatality rate of roughly 2 percent as compared with a fatality rate of around 0.1 percent for contemporary strains of influenza virus (IOM, 2004). The 1918 pandemic also exhibited an unprecedented excess of mortality among people from 20 to 40 years old, a population that has historically demonstrated a low influenza mortality rate, both in the years prior to 1918 and the years since (Morens and Fauci, 2007). Military populations in particular were severely affected by the 1918 pandemic.
The genome of the 1918 influenza virus has been recently sequenced from RNA fragments obtained from archival and frozen lung tissue of viral pneumonia victims. From that analysis the extinct pandemic strain of 1918 was identified as an influenza A virus, subtype H1N1. Studies under high-containment conditions using the reconstructed 1918 virus showed a high fatality rate in animal models 3 to 4 days after infection, a characteristic not described for any other human influenza virus (Tumpey et al., 2005).
The two other major influenza pandemics of the twentieth century were milder and less devastating than the 1918 pandemic. The first of these was the Asian flu pandemic of 1957-1958 caused by subtype H2N2. The virus, first found in the Far East, was identified in the United States in June 1957, with the first outbreak occurring in September and the peak coming in October. The timing of the epidemics coincided with the opening of the winter school term (Payne, 1958). A second wave hit in early 1958. Fatalities occurred mainly among the very young and the very old and in total claimed the lives of roughly 1,000,000 people (Potter, 2001).
The pandemic of 1968, caused by an H3N2 strain descended from H2N2 by antigenic shift (genetic reassortment), was milder than the 1957 pandemic, yet it still was estimated to have caused 500,000 deaths worldwide. The new strain was first isolated in Hong Kong in July 1968, although it most likely originated in nearby Guangdong Province, China. The pandemic proved less deadly in the United States than expected. Experts have suggested a variety of possible reasons: (1) antigenic overlap with the neuraminidase of the earlier H2N2 dominant strain, which would have provided a degree of cross protection; (2) the presence of preexisting antibodies (and protection) among the elderly born prior to 1893 as a result of the 1889-1891 pandemic; (3) the fact that the pandemic peaked around the holiday season in December, a time when most children are temporarily out of school, which would have slowed the spread of the virus; and (4) improvements in medical care and access to effective treatment (Dowdle, 1999; Kilbourne, 2006).
In 1977, an H1N1 virus reemerged in the Far East, genetically closely related to strains prevalent in the late (~1950) H1N1 era (Nakajima et al.,