INFLUENZA

Influenza (flu) is an important cause of morbidity and mortality. Flu is caused by infection with the influenza virus, a member of the Orthomyxoviridae family. Infection leads to illness, typically lasting a week, characterized by fever, sore throat, cough, headache, runny nose, and fatigue. Mild cases of flu can result in the common cold. In the medically more severe cases of flu, and in all fatal cases, a secondary pneumonia arises as a complication.

Flu is spread from person to person through sneezing and coughing. Influenza is zoonotic, that is, it has an animal origin, with avian and porcine strains able to jump species. The flu genome consists of eight single strands of RNA, which means that new strains can arise whenever an individual is infected with two existing strains. To be effective, flu vaccines, formulated on the basis of surveillance of early cases, must be given every year because of these constantly-changing strains.

Influenza is an epidemic disease in several respects. There is great year-to-year variability in morbidity, mortality, and incidence, and more intense periods are labeled as epidemics. New strains spread in epidemic style across regions, documented in a rich body of work by medical geographers, using both historical and modern data. There are occasional pandemics–severe outbreaks, global in scope; twentieth century pandemics occurred in 1918, 1946–1947, 1957, and 1968.

Age Profiles of Mortality

Figure 1 shows the age profile of death rates (per 100,000 population) for influenza and pneumonia (combined) for males and females in the United States in the years 1900, 1918, 1939, and 1998. Since fatal cases of influenza involve pneumonia, it is customary for statistical bureaus to merge influenza and pneumonia in published vital statistics. The patterns in the figure illustrate notable aspects of influenza demography and yield insight into mortality patterns more generally. To permit comparisons, all four panels in the figure are drawn to the same scale, with a horizontal rule across each panel at a mortality level of 100 per 100,000.

The influenza mortality rates exemplify three major mortality age patterns, named after letters of the alphabet: U, W, and J. In 1900, the pattern is U-shaped (sometimes called V-shaped), with peak

FIGURE 1

mortality at the upper and lower ranges of the age distribution. Though influenza occurs at all ages, mortality is concentrated among the youngest and oldest. A similar pattern is seen in 1939, except that the base of the U (but not the top prongs) has descended to a lower level and remains below the 100 per 100,000 line until a much later age. The 1939 panel represents the end of the pre-antibiotic era. Flu, being a viral disease, is not treatable with antibiotics, but secondary pneumonias often involve or are exacerbated by bacterial coinfection, which can be treated with antibiotics.

The mortality pattern in 1918 is completely atypical, even for a pandemic. Due to that year's epidemic of hypervirulent influenza, the pattern is W-shaped, with a peak at middle age in addition to peaks at infancy and old age. Such a pattern is unusual among biological causes of death: tuberculosis is perhaps the closest parallel. The entire mortality curve in 1918 lies above the 100 per 100,000 line, reflecting the severity of the epidemic. The male excess death rate, in absolute terms, was also exceptionally high in 1918, particularly at the middle-age peak of the W-pattern. The leading explanation for the decline in death rates above age 35, which gives rise to the W shape, is that at least one flu strain that was circulating in the late nineteenth century was similar to the 1918 strain, and had imparted partial immunity to those who, by 1918, were above age 35; in demographic terms, it is a cohort-effect explanation.

Since 1950, influenza death rates in developed countries have declined more at young ages than among the elderly, transforming the U shape into a J shape, as seen in the 1998 data.

The 1918 Epidemic

The 1918 influenza epidemic was an important event not only in the history of influenza, but also in world demographic history. Usual estimates are 20 to 40 million deaths worldwide, with great uncertainty because much of the world did not have vital registration in 1918. By the measure of its death toll and its global reach, it was the biggest epidemic of any disease in the twentieth century. In the United States, the estimate is 550,000 deaths (0.5% of the population), with about one-fourth of the population having had recognizable cases of flu. The 1918 epidemic is sometimes called the "Spanish flu." It is not thought to have had a Spanish origin, but since Spain, as a neutral power, did not censor newspapers during World War I, early accounts of the flu came from Spain. There is debate about whether the epidemic began in the United States and then spread to Europe and the rest of the world, or vice versa. Late twentieth and early twenty-first century studies, using modern laboratory techniques and preserved tissue samples, have focused on deciphering the molecular basis for the virulence of the 1918 strain, so far without conclusive results.

There is a hypothesis, based on data from the United States, that the 1918 influenza epidemic had a selective effect. Specifically, the suggestion is that many of those who died of flu in 1918 already had chronic illness (most notably, tuberculosis, highly prevalent at the time). The supporting evidence is that after 1918, death rates, especially from tuberculosis, dropped relative to their pre-1918 trend. The logic is that if the 1918 victims had already been sicker than average, then the post-1918 population would be healthier and therefore death rates should decline. It is rare to see such selection effects in the population at large, but the magnitude of the 1918 epidemic was large enough to be exploited as a natural experiment (i.e., a natural event which mimics an actual experiment in certain respects).

As noted, there has been little reduction in influenza and pneumonia death rates above age 80, in contrast to younger ages, which have seen dramatic declines. With the worldwide population aging, influenza is certain to remain an important cause of death for many years to come. Moreover, the possibility of another pandemic of hypervirulent influenza, comparable to that of 1918, cannot be ruled out.

See also: Diseases, Infectious; Epidemics.

bibliography

Crosby, Alfred W. 1989. America's Forgotten Pandemic: The Influenza of 1918. Cambridge, Eng.: Cambridge University Press.

Kilbourne, Edwin D. 1987. Influenza. New York: Plenum.

Kolata, Gina. 1999. Flu: The Story of the Great Influenza Pandemic of 1918 and the Search for the Virus That Caused It. New York: Farrar, Strausand Giroux.

Noymer, Andrew, and Michel Garenne. 2000. "The 1918 Influenza Epidemic's Effects on Sex Differentials in Mortality in the United States." Population and Development Review 26(3): 565–581.

Pyle, Gerald F. 1986. The Diffusion of Influenza: Patterns and Paradigms. Totowa, NJ: Rowman & Littlefield.

Andrew Noymer